Full text of "A text-book of inorganic chemistry for university students"

See other formats

75 70 65

s(b

7J5 7JO 6^5

eb 55 so 45 4b

75 7p 6^5

6^0 5*5 5p 4!5 4(0

II |

75 70 6J5

eb 55 sb 45 4o

7J5 7JO 6,5

6|0 5^5 5|0 4J5 40

B;O

* r|o e;5

6|0 5J5 5p 45 4JO

, ,1

7,5 7p 6,5

6'p 5^ 5(3 4^ . 4P

7J5 7b 65

6|0 5J5 50 4|5 4(0

8,0

7*5 7^0 6^

60 5J5 5(3 4J5 4)0

7|5 7|0 6J5

6^0 5J5 5|0 45 4^>

7'5 7JD 6^

G|O 55 5)3 4J5 4p

.1!

II ^ 1

TJs 70 e|s

6,0 55 5|o 4J5 40

1 1

6p 5,5 50 4,5 4,0

11 ill f

— PoH — —

60 55 50 45 4b cm

— W >-t fironn »-< R1iir>--».4 TnHio-n >^ -Violet ^ •

Yellow:
Chart of Spectra, showing positions of Principal Lines.

A TEXT-BOOK OF
INORGANIC CHEMISTRY

MACMILLAN AND CO., LIMITED

LONDON . BOMBAY . CALCUTTA . MADRAS
MELBOURNE

THE MACMILLAN COMPANY

NEW YORK . BOSTON . CHICAGO
DALLAS . SAN FRANCISCO

THE MACMILLAN CO. OF CANADA, LTD.

TORONTO

A TEXT-BOOK OF
INORGANIC CHEMISTRY

FOR UNIVERSITY STUDENTS

J. R. PARTINGTON, M.B.E., D.Sc.

Professor of Chemistry at the East London College, University of
London ; late Fellow of Manchester University

MACMILLAN AND CO., LIMITED
ST. MARTIN'S STREET, LONDON

1921

PREFACE

THE present text-book, as its title indicates, is primarily intended
for students who have completed an introductory course of Matricu-
lation standard, although the more elementary parts of the subject
are included so as to make the book complete in itself. It is not
written for any particular examination, but should meet the
requirements in Inorganic Chemistry of students preparing for the
examinations of the Intermediate and Pass B.Sc. of British
universities. Brief accounts of technical processes and the
elements of Physical Chemistry are included, with worked examples
on the latter.

The Atomic Theory and the Periodic Law have been given
prominence, since their neglect unfailingly leads to obscurity and
triviality. In explaining the foundations of the Molecular Theory
I thought it desirable to deviate from the current practice of
referring atomic weights to the standard O = 16. In my own
experience, which is, I believe, that of most teachers, students have
sufficient difficulty in reaching a clear understanding of Avogadro's
Law without the additional burden of an illogical change of units
halfway through the argument. Since there was no obvious
necessity to introduce the unit O=16 at a later stage, I refrained
from doing so and referred atomic weights to H=l. The table
on p. 145 contains all the atomic weights on both standards. Unless
specially stated, all atomic weights given in the book are on the basis
H=l.

Summaries of chapters have been added where they seemed
likely to be useful in affording assistance in revision, and examples
on all chapters are provided. The student will do well to supplement

vi PREFACE

the numerical questions by additional examples from one of the
many text-books on chemical calculations.

Limitations of space prevented more than a bare mention of most
of the so-called " Rare Elements," many of which are now of great
importance in chemical industry and form part of articles familiar
in everyday life. Their chemical properties are also in many
cases of unusual interest.

A short account of Werner's theory is given, since the classical
theory of Valency, which is of fundamental importance in the
somewhat monotonous uniformity of the chemistry of carbon,
proves inadequate when any but the very simplest compounds of
the remaining elements are under consideration.

The last chapter is intended to be no more than an outline :
greater detail in this field would have been inconsistent with the
scope of the book, and even undesirable in the present somewhat
mobile state of the frontiers of this new knowledge.

It is of the utmost importance that students of Chemistry should
have opportunities of examining as many as possible of the sub-
stances referred to in text- books and that lectures should be
experimental. Practically all the experiments described are shown
in the lecture courses at East London College, and the teacher
will find no difficulty in supplying the details of manipulation,
which through lack of space could not be given in full. Students
should realise that descriptions in text-books are necessarily incom-
plete and need to be supplemented by an acquaintance with the
substances themselves. The spurious character of knowledge
imparted by mere blackboard methods is painfully familiar to all
examiners.

In consulting original sources, reading the proofs, and preparing
the index, I have had valuable assistance from my wife. Sir
Richard Gregory and Mr. A. T. Simmons have throughout placed
their experience freely at my disposal and given me every possible
assistance whilst the book was passing through the press. To all
who have afforded me help in these and other ways I tender my
sincere thanks.

To avoid the multiplication of material the Publishers have
allowed me to make use of several illustrations from other books.
In this connection the " Treatise on Chemistry " of Roscoe and
Schorlemmer ; Lowry's " Historical Introduction to Chemistry "
(which may be referred to for fuller details on the historical side) ;

PREFACE vii

Donington's " Classbook of Chemistry " ; Miers's " Mineralogy " ;
and Tutton's " Crystallography " may be specially mentioned.

The physical properties of substances (densities, boiling points,
etc.) have been compiled from the most recent sources, in the
hope that the book may also prove useful for reference purposes.

J. R. PAETINGTON.

EAST LONDON COLLEGE,
UNIVERSITY OF LONDON.
July, 1920.

CONTENTS

CHAPTER I

PURE SUBSTANCES AND MIXTURES

PAGE *

CHAPTER II

ELEMENTS, COMPOUNDS, AND SOLUTIONS • . . . 18

CHAPTER III

THE COMPOSITION OF THE AIR AND THE THEORY OF COMBUSTION 35 '

CHAPTER IV

THE COMPOSITION OF WATER. . , . 51

CHAPTER V

THE PHYSICAL PROPERTIES OF GASES AND VAPOURS . . 66

CHAPTER VI

SOLUTIONS AND THE PHASE RULE 91

CHAPTER VII

THE LAWS OF STOICHIOMETRY 110

CHAPTER VIII

THE ATOMIC THEORY 125 zL

CHAPTER IX

AVOGADRO'S HYPOTHESIS AND THE MOLECULE . . . . 138 -

CHAPTER X

OXYGEN 159

ix <^

x CONTENTS

PAGE

CHAPTER XI

HYDROGEN 180

CHAPTER XII

WATER 200

CHAPTER XIII

COMMON SALT. HYDROCHLORIC ACID. CHLORINE . . . 218

CHAPTER XIV

VALENCY AND THE STRUCTURE OF COMPOUNDS . . . 245

CHAPTER XV

THE MOTION OF MOLECULES 258

CHAPTER XVI

ELECTROLYSIS 274

CHAPTER XVII

THE MOLECULAR WEIGHTS OF SUBSTANCES IN SOLUTION . 299

CHAPTER XVIII
OZONE " 320

CHAPTER XIX

HYDROGEN PEROXIDE 333

CHAPTER XX

CHEMIC^T, EQUILIBRIUM, AND THE LAW OF MASS-ACTION . 344

CHAPTER XXI

THE OXIDES AND OXY-ACIDS OF CHLORINE .... 368

CHAPTER XXII

THE HALOGENS 393

CHAPTER XXIII

ATOMIC HEATS AND ISOMORPHISM

CHAPTER XXIV

THE CLASSIFICATION OF THE ELEMENTS AND THE PERIODIC LAW 450

CONTENTS xi

PAGE

CHAPTER XXV

SULPHUR AND ITS COMPOUNDS WITH HYDROGEN AND HALOGENS 473

CHAPTER XXVI

THE OXYGEN COMPOUNDS OF SULPHUR ..... 490

CHAPTER XXVII

SELENIUM AND TELLURIUM . . ... . . . 528

CHAPTER XXVIII

NITROGEN AND ITS COMPOUNDS WITH HYDROGEN . . . 535

CHAPTER XXIX

THE OXIDES AND OXY- ACIDS OF NITROGEN . . . .561

CHAPTER XXX

THE INACTIVE ELEMENTS 598

CHAPTER XXXI

PHOSPHORUS , 607

CHAPTER XXXII

ARSENIC AND ITS COMPOUNDS . 644

CHAPTER XXXIII

CARBON AND THE HYDROCARBONS 658

CHAPTER XXXIV

OXYGEN COMPOUNDS OF CARBON, ETC ** . 686

CHAPTER XXXV

BORON AND SILICON . 732

CHAPTER XXXVI

. SPECTRUM ANALYSIS 755

CHAPTER XXXVII

METALS AND ALLOYS 764

CHAPTER XXXVIII

THE METALS OF THE ALKALIES 770

xii CONTENTS

CHAPTER XXXIX

COPPER, SILVER, AND GOLD 804

CHAPTER XL

THE ALKALINE EARTH-METALS 838

CHAPTER XLI

THE METALS OF THE ZINC GROUP 854

CHAPTER XLII

VOLTAIC CELLS 879

CHAPTER XLIII

THE METALS OF GROUP III OF THE PERIODIC SYSTEM . . 890

CHAPTER XLIV

THE METALS

OF THE FOURTH GROUP

. 911

THE METALS
THE METALS

CHAPTER XLV

OF THE NITROGEN GROUP ....

931

CHAPTER XLVI

OF THE SULPHUR GROUP

. 946

MANGANESE
IRON .

CHAPTER XLVII

960

CHAPTER XLVIII

. 972

COBALT AND

CHAPTER XLIX

NICKEL

. 998

CHAPTER L

THE PLATINUM METALS

. 1006

CHAPTER LI

THE RADIO -ELEMENTS AND THE STRUCTURE OF THE ATOM

. 1016

ANSWERS TO
INDEX

EXAMPLES

. 1040
1043

INORGANIC CHEMISTRY FOR
UNIVERSITY STUDENTS

CHAPTER I

PURE SUBSTANCES AND MIXTURES

Different kinds of solid bodies. — Different materials may be dis-
tinguished from one another by their properties, the most obvious
of which is the physical state : solid, liquid, or gaseous. Many dif-
ferent bodies having the same physical state, however, may easily
be distinguished from one another. Thus, coal, sugar, and salt are
obviously three different solid bodies ; water and paraffin oil are
different liquids, and coal gas and atmospheric air are different
gases. These differences we express by saying that the bodies
differ in composition.

In beginning the study of Chemistry, we meet with a large
number of new substances, and the already large number of different
bodies known to us in common life appears to be greatly increased.

Thus, if we examine specimens of the following solids we find that
each has some characteristic colour, which enables us to pick it
out from the others :

Blue vitriol : blue. Potassium dichromate : bright red.

Green vitriol : light green. Chrome alum : dark purple.

Nickel sulphate : bright green. Potash alum : colourless.
Cobalt nitrate : purplish-red.

All the above solids occur in pieces with definite shapes, called
crystals, bounded by plane faces meeting in sharp edges. Even
if the colour of the solid is not characteristic, we can often distin-
guish the material by its crystalline form. Thus, the following
colourless solids have characteristic shapes :

Alum : octahedra (Fig. 1). Rock salt : cubes (Fig. 4).

Nitre : long crystals (Fig. 2). Potassium chlorate : scaly crys-

"Hypo": beady crystals tals (Fig. 5).

(Fig. 3). Washing soda : lumpy crystals

(Fig. 6).

B B

INORGANIC CHEMISTRY

CHAP.

Another distinguishing property is the density of the solid.

Lead nitrate, although it crystallises in the same form as alum,
is much heavier.

Solids often occur in the forms of powder
or lumps, and the crystalline form, although
often present in the grains of powder and
recognisable under the microscope, may
sometimes be absent altogether, even in large
pieces. Such solids are said to be amorphous,
as distinguished from crystalline. The frag-
ments obtained on breaking crystals have sharp
edges and plane faces, or show a crystalline
fracture ; whereas the fractured pieces of an
amorphous solid, such as glass or pitch, show
curved faces like the inside of a shell, and
hence are said to exhibit a conchoidal fracture.
Powders of the same colour may often be

distinguished by their different densities :

FIG. 1.— Alum Crystal.

White.

Heavy : Barium sulphate
Light : Magnesium carbonate

Red.

Mercuric oxide
Ferric oxide

Black.

Manganese dioxide
Charcoal

FIG. 2.— .Nitre Crystal.

FIG. 3.— Crystal of Sodium Thiosulphate
<" Hypo ").

A third method of distinguishing between different solids is by
the solubility in liquids, such as water. Thus, if finely powdered
lead nitrate and barium sulphate, both heavy, white powders, are
separately stirred up with hot water in beakers, the former passes
into solution, whilst the latter remains undissolved.

Solids when heated usually melt at characteristic temperatures
called their melting points. Nitre melts at 345°,* potassium
chlorate at 350°, " hypo " at 48°, and rock salt at 815°. Barium
sulphate melts only at a very high temperature, about 2000°,
whilst charpoal has never yet been fused.

* Temperatures throughout are in degrees Centigrade.

i PURE SUBSTANCES AND MIXTURES 3

Different kinds of liquids. — The existence of different varieties
of liquids may be appreciated by examining specimens of the
following :

Water : colourless, odourless,
boiling point 100°, density 1.

Alcohol : colourless, spiritu-
ous odour, boiling point 78 -3 °,
density 0-79.

Ether : colourless, strong
sweet odour, very light and
mobile, boiling point 35-6°,
density 0-73.

Sulphuric acid : colourless,
oily, heavy, boiling point 330°,

density 1-85. FIG. 4.— Rock-salt Crystal

Bromine : dark red, suf-
focating odour, very heavy, boiling point 60°, density 3-1.

Mercury : very heavy, opaque liquid, with metallic lustre, boil-
ing point 357°, density 13-6.

In this table we recognise the boiling point as a criterion of the com-
position of the liquid ; this is definite under a fixed pressure, say
atmospheric, for a pure liquid. The freezing point, or temperature

FIG. 5.— Crystal of Potassium Chlorate. FIG. 6.— Crystal of Washing Soda.

of solidification, of the liquid is another property which may be
used for its identification. Water freezes at 0°, bromine at

-7-5°, mercury at -394°, alcohol at -111-8°, and ether at -113°.

Different kinds of gases. — The existence of different kinds of
gases was not clearly recognised until the eighteenth century,
when Joseph Priestley showed that there were several gases
differing from atmospheric air in their properties.

The differences may be appreciated by comparing jars con-
taining the following gases : oxygen, hydrogen, carbon dioxide,
nitric oxide, and chlorine.

B 2

4 INORGANIC CHEMISTRY CHAP.

By simple observation we find that chlorine has a greenish -
yellow colour, whilst the other gases are colourless. These colourless
gases may, however, be distinguished by appropriate experiments.

EXPT. 1. — Remove the glass plates from the jars so as to bring the
gases in contact with the air. Nothing occurs except with the nitric
oxide, which produces deep red fumes.

EXPT. 2. — Pour a little lime water into the other jars, and shake.
The lime water is unchanged in appearance in all the jars except that
containing carbon dioxide, in which it becomes turbid and white.

EXPT. 3. — Insert a lighted taper into each of a new set of jars of the

FIG. 7.— Experiment with Carbon Dioxide.

In oxygen it burns with a brilliant flame, in chlorine with a
smoky red flame, but in the other jars it is extinguished. The hydrogen
itself, however, takes fire and burns with a pale blue flame.

EXPT. 4. — If a jar of carbon dioxide is held over a counterpoised
beaker on a sensitive balance, and slowly inverted so as to pour the gas
into the beaker (Fig. 7), the latter sinks, showing that carbon dioxide
is heavier than air, and has passed into the beaker. A taper inserted
into the beaker is extinguished.

If a jar of hydrogen is opened, mouth downwards, under an inverted
counterpoised beaker (Fig. 8), and slowly inclined so as to pour the gas
upwards into the beaker, the latter rises, showing that hydrogen is
lighter than air.

i PURE SUBSTANCES AND MIXTURES 5

Thus, gases differ in density, colour, combustibility, capacity
to support combustion, and action on lime water.

Solids and liquids have been distinguished by : colour, form
(solids), density, smell, melting point, freezing point, boiling point,
and solubility.

Many of these properties may be measured quantitatively,
and so the differentiation of the substances rendered more exact.
In general, it is sufficient to examine only a few properties in order
to identify the material ; thus, water and formic acid, although
boiling at nearly the same temperature, have different densities,

FIG. 8.— Experiment with Hydrogen.

1-00 and 1-23, respectively, and different freezing points, 0° and
8-3°. Formic acid also has a pungent acid smell.

Pure substances. — Crystals of the same solid, say copper sulphate,
usually differ considerably in size, often in shape, yet we should
say that all pieces of this material are composed of the same pure
substance ; in other words, so far as composition is concerned, we
take no account of accidental circumstances such as size or shape.
Again, there are two kinds of phosphorus known in commerce,
viz., white and red phosphorus, which differ entirely in appearance
and properties. Although these consist of the same material,
phosphorus, they are two different substances, since each has
specific properties, by means of which it may be recognised.

The group including rock salt, salt from brine springs, purified

6 INORGANIC CHEMISTRY CHAP.

sea salt, and chemically pure salt made in the laboratory, com-
prises materials which agree in their properties, apart from size
and shape (which we regard in chemistry as non-essential), and we
say that all the members of this group are composed of one substance,
common salt.

./Chemistry. — The possibility of arranging materials into groups
of definite substances reduces the apparent complexity and scope
of their study, because a large number of individual bodies may
belong to one group, i.e., be composed of the same substance.
The fact that bodies may be arranged in such groups is the funda-
mental law of Chemistry. Descriptive Chemistry may be defined as the
science which deals with the preparation and properties of substances, and
the relations which exist between them.

In some cases difficulty may arise in defining the properties of
bodies, with the view of placing them in groups of substances.
Thus, a piece of granite has different properties in different parts.
Also, if we base our definition on identity of properties, we shall
apparently require an infinite number of groups to accommodate
all the possible liquids produced by adding salt to water in varying
proportions. These difficulties are removed by a closer study of
the cases in which they arise.

Homogeneous and heterogeneous bodies. — Bodies differ according
to the properties of their component parts. A body such that all
the portions into which it can be divided by mechanical means
possess identical properties is called a homogeneous body. Thus,
glass, water, and air are homogeneous bodies. All pure substances,
in the strict sense, are homogeneous bodies, but the converse, as
we shall see, is not true.

A body exhibiting different properties in different parts is
called a heterogeneous body. Thus, a piece of granite is readily
perceived by inspection to consist of an aggregate of three different
minerals.

One of these minerals is pink, opaque, and capable (though with
difficulty) of being scratched with a knife ; it is felspar. A second is
colourless, transparent, and too hard to scratch with a knife ; this is
quartz. The remaining mineral is in the form of thin grey, or black,
plates," which can be split by a knife into very thin leaves ; it is known
as mica.

Since the constituents of aggregates such as granite can be

I PURE SUBSTANCES AND MIXTURES 7

separated by mechanical means, heterogeneous bodies are often
called mechanical mixtures.

The separate parts of a mechanical mixture, or heterogeneous
body, are now called phases. Quartz, felspar, and mica are three
phases existing in granite. A mixture of ice and water consists of
two phases, whilst a homogeneous body, even if divided into
several parts in space, constitutes only a single phase.

It is not necessary that the parts of a heterogeneous body should
be so sharply differentiated as those making up a piece of granite.
Quartz crystals often occur which exhibit colouring in different
parts (" smoky quartz "), and the intensity of the brown colour, due to
impurities, may shade off from one part of the crystal to another.
Although we can break off two widely-separated parts of the crystal
which appear quite different, and thus satisfy ourselves that the whole
crystal is heterogeneous, it is very difficult to fix any position where
definite colour change occurs.

Since bodies exist in three states, the following types of mechanical
mixtures may exist :

(1) solid + solid (4) liquid + liquid (6) gas + gas

(2) solid + liquid (5) liquid + gas (7) solid + liquid + gas.

(3) solid + gas.

The ultramicroscope. — The definitions of homogeneous and
heterogeneous bodies given above are only relative. Thus, milk,
which may seem homogeneous to the eye, is readily seen under
the microscope to consist of transparent globules of butter-fat
floating in a nearly transparent liquid. In some cases hetero-
geneity which is not perceptible even by the microscope may be
revealed by the scattering of light.

If a few drops of a solution of gum mastic in alcohol are added to
water in a glass trough, and stirred, the resulting liquid appears clear,
to the eye, even with the aid of the microscope. But if a beam of light
from a lantern is passed through the water, before and after the mastic
has been added, it will be found that very little light can be seen passing
through the clear water, but that the water to which mastic has been
added shows the path of the light as a bright, cloudy beam. The same
effect is perceived when a ray of sunlight passes through dusty air ; in
this case the particles of dust may be seen floating about in the beam

An instrument making use of this principle is the ultramicroscope.
This (Fig. 9) consists of an ordinary high -power microscope with the
object-glass dipping into the liquid to be examined, contained in
a small glass cell. A powerful beam of light, from the sun, or
an arc-lamp, is brought, by means of a lens, to a focus in the liquid

INORGANIC CHEMISTRY

CHAP.

lying just under the microscope. The presence of suspended par-
ticles in the liquid is then revealed by the light scattered from
them, and they appear as bright specks.

Whilst microscopic visibility ceases with particles of diameter
about 1-5 x 10~5 cm., or 0-15/*. (/x = 0-001 mm.), the ultra-
microscope reveals particles down to 5 x 10"7 cm., or 5/x/*
(fjifji = 10~6 mm.), or about one-hundredth the wave-length of
visible light, which is 4 X 10~5 cm. in the case of violet light,
and 8 X 10~5 cm. in the case of red light.

By the action of phosphorus on a solution of gold chloride,
ruby-red, apparently clear, solutions are obtained. These, under
the ultramicroscope, exhibit particles, which have been shown to
be about 5/x/x in diameter. Suspensions of this kind, containing
ultramicroscopic particles, are called colloidal solutions. Still
smaller particles of gold, not visible even with the ultramicroscope,
can act as nuclei, or centres of condensation, for the production of

ultramicroscopic particles,
and the diameter of these
nuclei has been estimated
at 10~7 cm. Zsigmondy, the
inventor of the ultramicro-
scope, therefore distin-
guishes three kinds of small
particles in liquids : microns,
microscopically visible,
diameter 10~3 to 10~5 cm.
(ordinary suspensions) :
FIG. 9.— Diagram of Ultramicroscope. submicrons, ultramicro-

scopically visible, diameter

10~5 to 5 x 10~7 cm. (colloidal solutions) ; amicrons, invisible,
but act as nuclei, diameter 10~7 cm. {colloidal solutions).

If a few drops of a solution of the red dye eosin be added to a
trough of water, through which a beam of light is passed, the path
of the beam is rendered visible by a beautiful green light, which is
not unlike the haze obtained with the mastic, except that it is
coloured. Under the ultramicroscope, however, no particles can
be detected, and the phenomenon is quite different from that
shown by turbid liquids ; the effect is known as fluorescence. The
two effects are readily distinguished by the fact that the light from
turbid media is polarised, whilst fluorescent light is not.

Matter may also be produced in the form of very thin films, of
the same order of thickness as the ultramicroscopic particles.
Thus, gold-leaf is beaten out to a thickness of only 10~5 cm., and
if a piece of burning magnesium ribbon is held behind a piece of
gold-leaf pressed between two sheets of glass, the gold is seen to
be translucent, and to let through a green light.

I PURE SUBSTANCES AND MIXTURES 9

Atoms. — We shall see later that there is a good deal of direct
evidence that all kinds of matter are made of exceedingly small
particles called atoms, which have diameters of the order of 10~8 cm.
These cannot be seen even by the ultramicroscope, but are brought
into evidence in other ways when X-rays are used instead of
ordinary light. These X-rays differ from light only in having
a much smaller wave-length, of the order of 10~8 cm., and if they
are allowed to fall on a crystal, such as rock salt, the effect pro-
duced is similar to the colour phenomena seen with visible light
falling on thin films such as soap-bubbles. The X-rays are not
visible, but the effect can be detected, and it indicates that the
crystals are made up of layers of atoms, separated by distances
of the order of 10~8 cm. (Chap. LI).

Thus, in reality, all kinds of matter are heterogeneous, since
they are aggregates of atoms. In practice, we limit the use of the
word heterogeneous to bodies seen by the ultramicroscope to
consist of different parts, and speak of other bodies as homo-
geneous.

The separation of the constituents of mixtures of solids.—
The separation of the phases of a mixture of solids may be effected
in many different ways.

(1) Mechanically, by picking out the different bodies, if the system
is sufficiently coarse-grained.

(2) By differences of density, say by stirring up a powder with
an inert liquid, the density of which lies between that of one of
the solids and those of the others. Thus, in powdered granite, the
minerals have the following densities : felspar, 2-57 ; quartz,
2-65 ; mica, 2-85. Hence, if the powder is shaken with a mix-
ture of density 2-6, composed of the liquids benzene, density
0-874, and methylene iodide, density 3-33, the felspar will float,
whilst the mica and quartz will sink. The two heavier minerals
may then be separated by another mixture of the liquids of
density 2-7. *

(3) By magnetism ; iron filings may be separated from admixture
with flowers of sulphur by their attraction, to a magnet, leaving the
sulphur behind.

(4) By electrification ; if a mixture of red lead and flowers of
sulphur is dusted on an ebonite plate rubbed with flannel, the
red lead, which becomes charged positively, adheres to the plate,
which is charged negatively, whilst the sulphur, which has the
same charge as the plate, does not adhere. If the plate is now
tapped gently on a sheet of paper, the sulphur with only a little red
lead falls off, leaving the red lead on the plate ; this may be
brushed off on to the paper, and the colours of the two powders
compared.

(5) By the different attractions of the solids for a liquid (surface

INORGANIC CHEMISTRY

CHAP.

tension) ; thus, if a mixture of powdered zinc blende (native
zinc sulphide) and sand is sprinkled on the surface of water, the
sand is wetted and sinks, but the blende is not wetted, and floats,
although it is heavier than water.

(6) By the different solubilities in a liquid ; if the mixture of iron
filings and sulphur is shaken with carbon disulphide the sulphur
dissolves, and the solution may be decanted from the iron, which
is insoluble. If the solution of sulphur is poured into a dish,
covered with a filter paper, and the solvent allowed to evaporate,
fine crystals of sulphur are left (Fig. 10).

(7) By fusibility ; if a mixture of lead shot and sand is heated
in a crucible, the lead fuses, and the sand floats to the top.

(8) By volatility; if a
mixture of sand and
sulphur is heated in a
test-tube, the sulphur
boils, giving a reddish-
brown vapour which con-
denses on the cool sides
of the tube as a yellow
sublimate, whilst the sand
is left in the bottom of
the tube. A mixture of
sand and iodine behaves
similarly, the iodine
forming a beautiful violet
vapour, which condenses
on the cool tube as a black
crystalline sublimate.

FIG. 10.— Sulphur Crystals.

The magnetic method
is used to separate minerals
such as tinstone (density

6-4-7-1) and wolfram (density 7-1-7-9), which occur together, and
are so nearly alike in density that they cannot be separated by
washing with a stream of water. Tinstone is non -magnetic, whilst
wolfram (an ore of tungsten, a metal used in making the filaments of
electric lamps) is fairly magnetic. The crushed ore is dropped on a
travelling belt (Fig. 11), and falls off near a powerful electromagnet.
The tinstone falls undeflected, but the wolfram is pulled towards the
magnet, and forms a separate heap. The process is called electro-
magnetic separation.

Separation by surface tension is used in »the flotation process for
separating minerals, such as zinc blende, which is not easily wetted by
water, from galena (an ore of lead), which is. The crushed ore is agitated,

I PURE SUBSTANCES AND MIXTURES 11

by a blast of air, with water, to which a little oil, e.g., of eucalyptus,
has been added. The blende forms a scum on the surface, whilst the
galena sinks.

The separation of solids from liquids. — Solids mixed with liquids
may be separated in various ways.

(1) By settling out under the influence of gravity, i.e., by
sedimentation. The coarser the particles, the more rapidly they settle.

The rate of settling of spherical particles, solid or liquid, in a liquid
or gas is given by Stokes' s equation :

c = - r—^ (d — d'} cm. per sec. ;
9 77

where r = radius of the particles in cm. ; g = accelsration of gravity,
981 cm. per sec. per sec. ; rj = viscosity of the liquid in C.G.S. units ;

FIG. 11. — Diagram of Electromagnetic Separation.

d and d', the densities of the suspended particles, and of the liquid,
respectively. (If d < d', the particles rise ; e.g., air bubbles in water.)

The viscosity of a liquid is a measure of the resistance encountered
in stirring it ; ether is a liquid of small viscosity, treacle one of great
viscosity.

If we calculate from Stokes' s formula the rates of deposition of par-
ticles of sulphur (d = 2-06) in water (d = 1-00; 17 = 11-4 X 10~3 at
15°), the diameters of the sulphur particles being 0-1 cm., and
0-0001 cm., we find these rates to be 203 cm. per sec., and
0-000203 cm. per sec., respectively. The fine particles remain almost
permanently in suspension, forming a colloidal solution (see p. 8).

Particles of different sizes mixed together may be separated
by fractional sedimentation ; the powder is stirred up with water,

INORGANIC CHEMISTRY

CHAP.

and the time of settling divided into a number of intervals.
A series of powders increasing in fineness is thus obtained. The
process may be repeated with each of these. This method is
used in separating fine clay from coarse earth,
for the manufacture of porcelain.

(2) By centrifugal force. — A centrifugal
machine is shown in Fig. 12. Two alu-
minium tubes are attached by hinges to a
central shaft, which may be rotated a.t high
speed (2000 revs, per min.) by means of
gearing and the handle. If a glass tube is
filled with the fine suspension of barium
sulphate obtained by adding dilute sulphuric
acid to a solution of barium chloride, and is
placed in one of the aluminium tubes, a
similar tube of water being put in the
opposite side as a counterpoise, the powder
is separated on the bottom of the tube on
working the machine. A comparison tube
of the suspension, kept at rest, does not
settle during the time of the experiment.

(3) In many cases suspended particles
are charged electrically, and move in an
electric field ; this motion is known as
cataphoresis.

EXPT. 5. — A colloidal solution of arsenic sulphide (i.e., a suspension of
very fine particles) is made by pouring a solution of arsenic trioxide in
boiling water into a solution of sulphuretted hydrogen
in water, and driving out the excess of the latter
gas by a stream of hydrogen. About 50 c.c. are taken,
and about 5 gm. of urea dissolved in it to make it
denser than water. The yellow solution is carefully
run by a pipette into the lower part of a U-tube half
filled with distilled water. Platinum plates fitted
through glass tubes by platinum wires are placed in
each arm of the tube, and connected with the supply
mains (220 volts). The level of the arsenic sulphide,
as marked by paper rings, soon falls on one side of the
U-tube, and rises on the other (Fig. 13). The fact
that the colloidal solution contains suspended particles
may be seen by passing a beam of light through some
of it in a beaker.

If a few drops of sulphuric acid are added to the colloidal solution
of arsenic sulphide, an immediate precipitation in yellow flocks
occurs. If the. mixture is now shaken with paraffin oil, the latter

Fm. 12.

Centrifugal Machine.

FIG. 13.
Cataphoresis.

i PURE SUBSTANCES AND MIXTURES 13

rises to the surface, carrying the yellow arsenic sulphide with it.
This is an application of the flotation process (p. 10) : arsenic sulphide
adheres more strongly to oil than to water.

(4) The commonest method of separating solids from liquids is
by filtration. The liquid containing the suspended precipitate
is poured on a filter, consisting of a folded cone of unglazed paper
in a glass funnel. The liquid passes through the pores of the
paper under the action of gravity, but the solid particles, if larger
than the pores of the paper, are kept back.

Particles which are very small pass through the filter. Thus, the
precipitate of barium sulphate prepared above runs as a milky liquid
through an ordinary filter paper. A special " barium sulphate paper, v

of fine texture, may then be used. water*

The size of the particles of this
precipitate may be increased by
precipitating a boiling solution of
barium chloride with boiling dilute
sulphuric acid. The barium sulphate
then settles out very rapidly, and is
easily filtered.

Hot solutions filter more rapidly
than cold ones, since the viscosity
of the liquid is reduced by raising
the temperature, and the process
of filtration is really the passage
of liquid through capillary tubes,
the speed increasing as the vis-
cosity diminishes.

The rate of filtration is also
increased by increasing the differ-
ence of pressure between the two ends of the capillary pores of the
filter. This is effected by filtration under reduced pressure. The
filter paper is laid flat on the perforated grid of a porcelain filter
funnel (Biichner funnel) (Fig. 14), which is fitted air-tight through
a rubber stopper into a 'tubulated filter flask. The side tube of
the filter flask is connected with a filter pump, actuated by a stream
of water from the mains. The air is removed from the flask,
and the pressure difference on the two sides of the paper thus
increased. Such filter funnels should not be allowed to become
empty during filtration and washing, as then air-channels are
formed in the precipitate. In washing precipitates in ordinary
funnels, on the contrary, each lot of liquid should be allowed to
drain out before the next is added.

The separation of liquids from liquids. — If chloroform and water

INORGANIC CHEMISTRY

CHAP.

are shaken together in a separating funnel (Fig. 15), and then allowed

to stand, the heavy chloroform settles out in a layer at the bottom,

and can be run off through the tap, leaving the water in the bulb

of the funnel.

A suspension of fine droplets of one liquid in another is called an

emulsion. Thus, milk is an emulsion of droplets of fat in a watery
liquid. Each liquid alone may be transparent, but
the emulsion is turbid, owing to the scattering of light
from the small particles.

Emulsions can often be separated by centrifugal
force ; milk is separated partially in this way into
cream (rich in fat), and separated milk (poor in
fat).

If one liquid is soluble in a third, whilst the
second is not, a separation may be effected by
shaking with the third liquid. If ether is added to
an emulsion of paraffin oil and water in a separating
funnel, and shaken with it, the ether dissolves the
paraffin, and the solution floats to the surface. The
water may be run off below, and the ethereal solution
allowed to evaporate on a water-bath (Fig. 16),
leaving the oil.

The separation of gases from liquids and solids.—
Gases mix with each other in all proportions, so that
heterogeneous systems can be
Separating obtained with gases only trans-
Funnel, iently, when a light gas is
stratified on a heavier one.

The line of demarcation is not sharp, and

the gases soon mix by diffusion.

EXPT. 6. — Pass carbon dioxide into a large
jar, so that the latter is partially filled with
the gas (Fig. 17). Blow a bubble with
Plateau's soap solution * and allow it to fall
into the jar. It is arrested on reaching the
carbon dioxide, and remains suspended. If
a taper is lowered into the jar, it is extinguished
on reaching the level of the bubble.

A suspension of minute bubbles of gas in a liquid forms a froth

* Plateau's Soap Solution is prepared as follows. 10 gm. of sodium oleate
and 400 c.c. of distilled water are allowed to stand at the ordinary temperature
in a stoppered bottle until solution occurs. 100 c.c. of pure glycerin are then
added, and the liquid, after shaking, is allowed to settle for a few days in the
dark. The clear liquid is decanted or siphoned off, and, after the addition of
1 drop of ammonia, is preserved in a stoppered bottle covered outside with
opaque black varnish.

Fia. 16.— Evaporation on a
Water-bath.

r PURE SUBSTANCES AND MIXTURES 15

or foam. It is usually produced by shaking the gas with a liquid of
low surface-tension, such as soap solution. Froths may be separated
by centrifugal force, or by adding other liquids, such as alcohol
to aqueous foams.

A suspension of minute droplets of liquid in a gas, such as is
produced by rapidly cooling moist air, is called a mist or fog. In
fogs the particles are smaller, and a mist may pass over into rain
when the particles of liquid unite by coalescing into large drops.

Aitkcn showed that mists are produced by condensation on
minute solid particles of dust (motes) floating in the air ; if these
are partially removed, say by filtering the air through cotton-
wool, then, on cooling, condensation occurs
on the few remaining nuclei, producing
rain-like drops. If the nuclei are all
removed, by allowing the air to stand for
some time in a vessel with wetted sides,
then condensation does not occur at all
until the air has been cooled much below
the temperature at which mist-formation
previously took place.

C. T. R. Wilson found that minute electrically
charged nuclei, called gaseous ions, which
are produced even in dust-free air by electric
sparks, or exposure to X-rays, can also act
as condensation centres. They may also be
filtered out by cotton-wool (Chap. LI).

A suspension of fine particles of solid in a

^ , FIG. 17.— Experiment illus-

gas is called a smoke or tume. Loal smoke trating stratification of
consists of small particles of carbon, which Gases-
when they aggregate together form soot.

The smoke from the glowing tip of a cigarette, which also consists of
small particles of carbon, appears blue, because the particles are very
fine, their diameters being of the order of a wave-length of light.
Smoke rising vertically from a chimney in clear dry air also appears
blue. The smoke some distance from the end of the cigarette, or
blown from the mouth, and smoke from a chimney on a damp day,
appear greyish-white and opaque, because the particles are larger,
probably as a result of the condensation of moisture upon them.

The particles of fogs and smokes are often electrically charged, or
become so on exposure to a high-tension discharge such as is given
off from a point or fine wire attached to a pole of an electrical
machine or induction coil. During such discharges, the fume is
often precipitated, as was shown by Sir Oliver Lodge in 1883.
This method of fume dissipation has recently been applied by

16 INORGANIC CHEMISTRY CHAP.

Dr. F. G. Cottrell, in America, to the precipitation of fumes from
smelting furnaces, blast-furnaces, cement-furnaces, etc.

EXPT. 7. — Fill a bell -jar with fumes of ammonium chloride by passing
air through two flasks containing strong hydrochloric acid and ammonia
solution, respectively (Fig. 18). Place the bell -jar on a metal plate
connected with one pole of an induction coil, or Wimshurst machine, and
connect the other pole with a pointed copper wire passing through a
rubber stopper in the bell -jar. On electrifying the apparatus, the fume
rapidly settles. A comparison plain jar filled with fume is placed beside
the first one, to show the persistence of the fume without treatment.

The Cottrell apparatus consists of tubes or chambers containing
electrodes, between which a high tension of 75,000 volts is main-

FIG. 18.— Electrical Fume Precipitation.

tained. The solid deposited from the fume 'passing through is
shaken off the sides of the tube or chamber from time to time by
tapping with an automatic hammer ; liquids flow away without
such treatment.

SUMMARY OF CHAPTER I

Different kinds of matter exist, characterised by different properties
when examined under the same conditions. Some masses of matter
are homogeneous, i.e., of the same kind throughout, whilst others are
heterogeneous, i.e., of different kinds in different parts of the mass. All
the parts of a heterogeneous mass may be separated from one an-
other by suitable means, depending on differences in density, magnetic
and electrical properties, surface-tension, solubility, volatility, fusibility,
etc.

PURE SUBSTANCES AND MIXTURES 17

EXERCISES ON CHAPTER I

1. Describe some of the means available for differentiating between
various kinds of solid bodies. If you were given two white powders,
one of which was silica and the other lead carbonate, how would
you determine which was which by non-chemical means ?

2. Tabulate the various methods used in the separation of mechanical
mixtures of : (a) solids and liquids, (6) solids and solids, (c) liquids and
liquids, (d) gases and liquids, pointing out methods common to the four
classes. In which classes of heterogeneous bodies would you place
(a) milk, (6) snow, (c) pumice-stone, (d) white paint ?

3. Explain what is meant by the terms : phase, heterogeneous,
homogeneous, colloidal solution, precipitate. Discuss the use of the
term heterogeneous as applied to matter in general.

4. Compare the rates at which particles of silica (density 2-65) of
diameters 0-25 and 0-01 mm., respectively, settle in water. How may
the rate of settling be accelerated ?

5. How are (a) flotation, (6) electric precipitation, (c) electromagnetic
separation, applied on the large scale ?

CHAPTER II

ELEMENTS, COMPOUNDS, AND SOLUTIONS

Chemical changes. — It is a matter of common observation that
bodies often undergo radical changes under certain conditions.
Thus, wine on standing exposed to air may lose its colour, and
become sour ; bright copper becomes dull, and ultimately covered
with a green crust, when exposed to moist air, and under the same
conditions iron rusts away completely to a brown powder. A
candle burns away, and apparently disappears.

In other cases the changes appear to be much less deep-seated,
and the properties of the materials are only slightly, and tem-
porarily, modified. Thus, water on cooling freezes to ice, but the
ice melts, and is reconverted into water, on warming. A bar of
iron which has been heated to redness is only slightly altered and,
apart from a little scale on the surface, is recovered without
change on- cooling.

EXPT. 8. — Heat in a Bunsen flame a piece of platinum wire. The
wire becomes red-hot, but on cooling is apparently quite unchanged.
Repeat the experiment with a piece of magnesium ribbon. The ribbon
takes fire and burns with a brilliant white flame, producing a white ash.

Material changes are found, by such observations and experi-
ments, to be divisible into two large but not sharply defined classes :
either they affect only a few properties of the material, and are
temporary, or they are much more drastic, resulting in the dis-
appearance of the original material as such, and the formation in
its place of a different material. Changes of the first class are
called physical changes ; those of the second class, chemical changes.

EXPT. 9. — Place a small piece of yellow phosphorus on a sand-tray,
and sprinkle over it a few crystals of iodine. The phosphorus takes
fire.

CH. ii ELEMENTS, COMPOUNDS, AND SOLUTIONS 19

EXPT. 10. — Pour into separate test-glasses a little of the following
solutions : potassium ferricyanide, tannin, potassium thiocyanate,
caustic potash. Add to each glass a dilute solution of ferric chloride.
A blue, black, blood-red, and brown liquid, respectively, is produced.

EXPT. 11. — Heat a small pill of mercuric thiocyanate by the flame of
a taper. The substance swells up into a worm-like mass of a friable
brown substance (" Pharaoh's Serpent ").

EXPT. 12. — Heat a mixture of 5 parts of fine iron filings and 3 parts by
weight of flowers of sulphur in a test-tube. The sulphur boils, and then
the iron begins to glow, and continues to do so when the tube is removed
from the flame. When the glowing ceases, heat the tube for a short
time, then allow it to cool by placing it on a tray of sand. When cold,
break the tube carefully in a mortar. A greyish mass is obtained, which
is easily powdered in the mortar. The powder is black, and under a
lens no iron or sulphur particles can be distinguished in it. It yields no
sulphur when treated with carbon disulphide (p. 10), and if a magnet is
brought over it, the powder is completely attracted (although it must
be removed in portions since it is not so magnetic as iron), leaving no
residue of sulphur, as was the case with the original mixture. The iron
and sulphur have formed a new substance, called iron sulphide.

From these experiments it is seen that chemical changes are
often accompanied by an evolution of heat. This, however, is
by no means always the case, since sometimes heat is absorbed.

EXPT. 13. — Pour concentrated hydrochloric acid over crystals of
Glauber's salt in a beaker. The crystals fall to a granular white
powder, which may be recognised, if filtered off, as common salt. A
considerable absorption of heat occurs, and the beaker feels very cold.
If a small test-tube of water is placed in the mixture in the beaker, the
water is quickly frozen.

The law of conservation of matter. — The quantity of matter in a
body is measured by its weight. The weight of a body, however, de-
pends on the force of gravity attracting the body to the centre of the
earth, and the latter varies slightly from place to place on the sur-
face of the earth. In the ordinary balance this slight variation
affects equally both the body weighed and the weights used in the
other pan, so that the weight appears always to be the same. If a
spring-balance is used, slight differences are found in different
localities, since the weight is then measured directly by the extension
produced in a spring by the attraction of gravity. The name mass
is therefore used to indicate the property of the body of resisting
the action of forces tending to set it in motion, one such force being
gravity. The mass of a body is supposed to be an unalterable

c 2

20 INORGANIC CHEMISTRY CHAP.

property of the body itself, and a measure of the quantity of matter
contained in the body.

The ancient philosophers had views on the ultimate fixity of the
material world. Thus, Empedocles (B.C. 490-430), as quoted by
Aristotle, says : " Nothing can be made out of nothing, and it is
impossible to annihilate anything. All that happens in the world
depends on a change of form and upon the mixture, or separation,
of bodies." This is strikingly similar to the statement of the
French chemist Lavoisier (A.D. 1743-1794), made about 2300 years
later : " Nothing can be created, and in every process there is
just as much substance (quantity of matter) present before and
after the process has taken place. There is only a change or
modification of the matter." Lavoisier's statement, however,
differed from that of Empedocles : whereas the statement of the
Greek philosopher was merely an unverified opinion, that of
Lavoisier was a scientific truth, founded upon experiment.

The early chemists, with one or two exceptions, entirely ignored
the changes of weight occurring in chemical processes. Usually
they considered such matters as removed from purely chemical
studies, and beneath their notice. Thus, Jean Rey (1630) says :
" The examination of weights by the balance differs from that
made by the reason. The latter is only employed by the Judici-
ous, whilst the former can be practised by the Veriest Clown. The
latter is always exact, whilst the former is seldom without deception."

Joseph Black (1755), in a research on magnesia, paid careful
attention to the weights of the materials. " Three ounces of
magnesia were distilled in a glass retort and receiver. When all
was cool, I found only five drachms of whitish water in the receiver
. . . the magnesia when taken out of the retort . . . had lost
half its weight ... It is evident that of the volatile parts con-
tained in the powder, a small portion only is water ; the rest cannot,
it seems, be retained in vessels under a visible form . . . and is
mostly air [carbon dioxide]." Thus, when Black found a loss of
weight in a chemical change, he put it down to the escape of some
material which had escaped attention, and he began to look for this
material. In doing this, he recognised implicitly the principle
stated later by Lavoisier. Black's experiment is an example of
many chemical changes in which an apparent destruction of matter
is due to the escape of a gas, which is very likely to be overlooked
unless special search is made for it. Since the existence of gases
was not 'recognised clearly until the eighteenth century, it is not
surprising that a belief in the actual destruction of matter should
have survived until that period.

Experiments on the conservation of matter.— When a candle
burns, it is apparently completely destroyed. It is easy to show by
experiment that this is not the case.

ELEMENTS, COMPOUNDS, AND SOLUTIONS

EXPT. 14. — Fit a small candle through a cork, in which there are
four holes bored to admit air, into a glass tube 2 in. wide and 8 in. long,
in which a piece of wire -gauze is supported by three wires from the top.
Sticks of caustic soda, supported on a few pieces of quicklime, are placed
on the top of the gauze, and the whole apparatus is counterpoised on
one arm of a balance (Fig. 19). Light the candle, and allow it to burn.
In a few minutes the arm of the balance carrying the apparatus is
depressed, showing that, so far from a loss of weight occurring when a
candle burns, there is an increase of weight if the products of combustion
are prevented from escaping by absorption in caustic soda. The nature
of these products may be found by the following experiments.

EXPT. 15. — Hold a dry bell -jar over

a burning candle. The sides of the

jar are at once

dimmed by mois-

t u r e deposited

upon the cold surface. Hence water
is one of the products of combustion
of a candle.

Burn a candle, supported by a wire,
in a gas jar. Pour a little lime-water
into the jar : on shaking, it becomes
turbid. Hence carbon dioxide is
produced by the combustion.

Both water and carbon dioxide are
retained by quicklime and caustic
soda.

The increase in weight in
Expt. 14 renders it probable that
the air has taken some part in
the combustion, and that the pro-
ducts of combustion, which are
absorbed by the caustic soda, con-
tain part of the air. If this is the case, air must possess weight.
Although the ancients believed that air was without weight, the
opposite was proved by the following experiment of Otto von
Guericke, the inventor of the air-pump (1650).

EXPT. 16. — Evacuate by an air-pump, and counterpoise on the
balance a 2 -litre globe, fitted with a stopcock through a rubber cork
(Fig. 20). Open the stopcock, and notice the hissing noise of the air
rushing into the globe. Replace the globe on the balance, and notice
that the side of the beam on which it hangs is now depressed.

In order to test the truth of Lavoisier's statement, it is obvious

FIG. 19.

Burning of

Candle.

FIG. 20.— Flask for
Weighing Air.

INORGANIC CHEMISTRY CHAP.

that the chemical change, or chemical reaction, as it is usually
called, must be instituted in a closed space, so that none of the
materials used can escape.

EXPT. 17. — Place a small piece of phosphorus, dried by pressing
between filter paper, in a dry strong round-bottom flask of about
250 c.c. capacity, fitted with a good rubber stopper. Weigh the flask.
Warm over a flame the spot where the phosphorus lies until it ignites.
When the combustion is finished, allow the flask to cool, and reweigh.
The weight is unchanged.

EXPT. 18. — Pour a little mercuric cliloride solution into a conical
flask, and place inside a small tube containing a solution of potassium
iodide. Cork the flask (Fig. 21), and counterpoise on the balance. Now
tilt the flask so that the solutions mix. A red
precipitate of mercury iodide is formed, but the
weight will be found to be unchanged.

The generalisation stated by Lavoisier is
called the Law of Conservation of Matter, or the
Law of Indestructibility of Matter. It is true both
for physical and for chemical changes. Some
very exact experiments have been made to
test the degree of accuracy of the law.
. Stas (1865) took 27-6223 gm. of pure
servation of Matter. silver, and 32-4665 gm. of pure iodine, and
by a roundabout series of chemical reactions
converted them into silver iodide, which weighed 60-0860 gm.
The loss of weight is only 0-0028 gm., or 0-00005 of the total weight.
E. W. Morley (1895) combined 30-3429 gm. of pure oxygen with
3-8211 gm. of pure hydrogen, and obtained 34-1559 gm. of water.
The loss of weight is 0-0081 gm., or 0-0002 of the total weight.

Landolt's experiments. — Until 1900 the law of conservation of
matter was regarded as accurate within the limits of experi-
mental error, which were very small. In that year, however,
Heydweiller stated that he had observed small losses of weight
when certain chemical reactions were carried out in sealed vessels.
Thus, when 80 gm. of copper sulphate, dissolved in 130 c.c. of water,
were decomposed with 15 gm. of metallic iron, there was a loss of
weight of 0-217 mgm. H. Landolt in 1893 began a series of
experiments, which were not completed until 1908, with the object
of testing the law of conservation of matter with all possible accuracy,
and of determining whether the deviations noticed were real, or
due to some error of experiment.

In the separate legs of a Jena glass U-tube (Fig. 22) Landolt sealed
up solutions of substances capable of reacting without the production

ELEMENTS, COMPOUNDS, AND SOLUTIONS

of much heat, so that the disturbances arising from this cause could be
eliminated. He used :

1. Silver sulphate and ferrous sulphate, giving a precipitate of metallic
silver.

2. Hydriodic acid and iodic acid, giving a precipitate of iodine.

3. Iodine and sodium sulphite, giving sodium iodide and sulphate.

4. Chloral hydrate and caustic potash, giving an emulsion of chloro-
form.

The tube was counterpoised against an exactly similar tube on a
balance capable (in the final experiments) of detecting a change of weight
of 0-0001 gm. with a load of 1 kgm. in each pan, i.e., a change of 1 part
in 10,000,000. One reaction tube was then inverted, after removing it
from the balance, and the chemical change allowed to take place. After
cooling, the tube was replaced on the balance, and the change in weight,
usually a diminution, noted. The other tube was
then taken off and inverted, and the process
repeated. At first, slight diminutions in weight,
amounting to 0-167 mgm. in the maximum, were
always found, but after a long series of experiments
these were traced to two causes : (a) as a result
of the slight evolution of heat, the film of moisture
condensed on the outer surface of the glass was
partially driven off, and did not return until after
long standing ; (6) the vessel expanded slightly
as a result of the slight increase of temperature,
and did not return to its original volume until
some time had elapsed. In consequence of the first
error, the weight of the vessel was reduced, and the
second error, which led to an increase m the volume
of air displaced by the vessel, also reduced the apparent weight. By
allowing the vessel to stand for a long time after the experiment, before
reweighing, Landolt found that it recovered its original weight to within
1 part in 10,000,000 — i.e., within the limits of experimental error. For
these reactions, therefore, the law of conservation of matter must be
considered to be an exact law. Whether it holds exactly for reactions
in which there is considerable evolution of heat, such as the combustion
of phosphorus, cannot be stated, since no experiments have been made
with sufficient accuracy in these cases.

Elements and compounds. — Homogeneous materials may undergo
chemical changes in one of two ways, according to their com-
position. Either the substance increases in weight in all the changes
which it undergoes ; or it gives other substances, each of less weight
than the original substance, or, as is said, decomposes.

FIG. 22.— Landolt's
Experiment.

24 INORGANIC CHEMISTRY CHAP.

EXPT. 19. — Heat 0-5 gm. of magnesium ribbon in a weighed, loosely
closed, porcelain crucible, with a small flame, till combustion ceases
(Fig. 23). Then heat strongly for ten minutes, cool, and reweigh.
There is an increase in weight, which, if the experiment is performed
carefully, amounts to 0-333 gm.

EXPT. 20. — Heat 2-16 gm. of red oxide of mercury in a weighed,
hard glass tube, connected by a rubber stopper with a glass delivery tube
leading to a pneumatic trough in which is inverted a 200 c.c. measuring
cylinder full of water, the mouth of which is over the delivery tube
(Fig. 24). The red substance blackens, and bubbles of gas collect in the
cylinder. At the same time, a shining metallic sublimate of mercury
collects on the cooler part of the hard glass tube, which is supported in
a horizontal position to prevent the globules of mercury which condense
running back on to the hot glass. When the evolution of gas ceases
and the oxide has disappeared, remove the delivery tube from the trough

and allow the apparatus to cool. Reweigh
the tube, and note that it has lost in
weight. If the experiment has been care-
fully performed the loss in weight should
amount to 0 • 1 60 gm. The volume of gas
collected will be about 118 c.c. If a
glowing chip of wood is placed in the gas,
it is rekindled, and burns with a brilliant
flame, indicating that the gas is oxygen.

If a pure substance can be decomposed
into two or more substances of smaller
weight, as the red oxide of mercury into
FIG. 23.— Heating Magnesium mercury and oxygen gas, we say that

it is a Compound. If it always yields

substances of greater weight, indicating that, in all reactions in
which it takes part, union always occurs with other substances,
and never decomposition into two or more substances, the sub-
stance is called an Element. Magnesium is an element.

At this point, however, we meet again with a difficulty en-
countered in Chapter I, viz., that in some cases a homogeneous
material may have a whole range of compositions according to
the way in which it is prepared. Solutions of common salt in
water may vary in composition from pure water to a liquid con-
taining 26-5 per cent, by weight of salt. Between these two
limits there is an infinite number of possible compositions. But
if we decompose red oxide of mercury, no matter how it has
been prepared (p. 26), we find that it always has the same

ELEMENTS, COMPOUNDS, AND SOLUTIONS

composition, containing 8 gm. of oxygen to 100 gm. of
mercury.

It is therefore necessary to divide into two classes all those homo-
geneous materials which are not elements. Those of constant
composition are called Compounds ; those of variable composition
are called Solutions. Red oxide of mercury is a compound, but the
liquids containing salt and water are solutions.

Solutions are sometimes called " Mixtures," but this name we
have reserved for heterogeneous systems, i.e., " Mechanical Mix-
tures," and it is therefore important to avoid confusion, by restricting
the use of the word.

Solutions are always separable, by suitable means, into two or

FIG. 24.— Decomposition of Oxide of Mercury by Heat.

more pure substances, either elements or compounds. Thus,
solutions of salt in water are separated into these two constituents
by simple evaporation.

In the above classification we remove the second difficulty
encountered in the definition of pure substances (p. 6). The homo-
geneous liquids formed from salt and water, for instance, are not
to be placed in separate groups of substances, the number of which
would then be infinite, but are to be regarded as solutions of two
pure substances, viz., salt and water, in varying proportions.

26 INORGANIC CHEMISTRY ^ CHAP.

We have now arrived at the following classification :

BODIES

]

Heterogeneous Homogeneous

systems of phases

Non-

Solutions Compounds Elements

(Variable composition) (Fixed composition)

Pure substances

Analysis and synthesis. — The process by which a compound is
separated into its constituent elements, e.g., the decomposition of red
oxide of mercury by heat, is called analysis, from the Greek
analuo, I unloose. The building up of a compound from its
elements, as in the production of magnesium oxide by heating
magnesium in air, is called synthesis, the Greek word synthesis mean-
ing a putting together. The process of ascertaining the compo-
sition of substances is also called analysis ; qualitative analysis
leads to a knowledge of the constituents only, without finding
the proportions in which they are combined, whilst quantitative
analysis determines these proportions in addition.

It follows from the definition of a compound that its composition
is independent of the method of preparation. The same compound,
also, gives the same elements in the same proportions, no matter
what means are used for its decomposition.

EXPT. 21. — Metallic tin may be converted into oxide of tin in three
different ways :

(a) One gm. of tinfoil is weighed into a counterpoised Rose's crucible
(Fig. 25), and heated in a stream of oxygen passed through the porcelain
tube through a small hole in the lid of the crucible. The crucible is
cooled and weighed from time to time until its weight becomes constant.
The residue is oxide of tin.

(b) One gm. of tinfoil is weighed into a counterpoised porcelain basin,
covered with a large watch-glass. It is treated carefully with suc-
cessive small amounts of strong nitric acid until the violent action

ii ELEMENTS, COMPOUNDS, AND SOLUTIONS 27

ceases, the watch-glass being placed over the basin after each addition
to prevent loss by spirting. The solid on the glass is washed into the
dish, and the excess of acid is then evaporated off on a sand-bath, and
the material heated for ten minutes over a Bunsen flame. The dish is
cooled and weighed. The residue is oxide of tin.

(c) One gm. of tinfoil is weighed into a conical flask and dissolved in
strong hydrochloric acid by warming on a sand-bath. The solution of
chloride of tin is diluted with water, and precipitated with a stream of
sulphuretted hydrogen. The tin sulphide is filtered and washed, and
the filter paper and precipitate ignited in a weighed porcelain crucible.
This is cooled and weighed. The residue is oxide of -tin.

It will be found that, within the limits of experimental error, the
weight of oxide of tin obtained from 1 gm. of tin in the three different
methods-is the same. Hence the composition of oxide of tin is constant,
and independent of the method of preparation.
Oxide of tin is, therefore, a compound, not- a mix-
ture or a solution.

The early history of chemistry. — The conceptions
underlying the definitions of elements and com-
pounds, although now almost obvious, were reached
only after centuries of effort. They represent
the few grains of truth remaining from the
winnowing process of experimental investigation
applied to the mass of opinions on the constitution
of bodies which had accumulated, either as
a heritage from antiquity, or from the equally FlG 95

unverified guesswork of the later alchemical Rose's Crucible.
period. It may therefore not be out of
place to. give a very brief account of the development of these
fundamental conceptions from the dawn of chemistry, one of the
oldest of the sciences.

The definition of an element given above dates from the seven-
teenth century, when Robert Boyle, in his " Sceptical Chymist "
(1661), agrees to use " elements and principles as terms equivalent,
and to understand both by the one and the other, those primitive
and simple bodies of which the mixed ones are composed, and
into which they are ultimately resolved." According to Boyle,
therefore, the elements are the practical limits of chemical analysis,
or are substances incapable of decomposition by any means with
which we are at present acquainted. This definition is provisional :
substances now regarded as elements may, at some future date,
with improved methods, be shown to be compounds, but until
that happens they must be regarded as elementary.

Theory of the four elements. — The first clear expression of the
idea of an element occurs in the teachings of the Greek philoso-

INORGANTC CHEMISTRY

CHAP.

pher, Aristotle (B.C. 384-322), who appears to have borrowed it
from earlier thinkers of antiquity. All substances were con-
sidered to be made of a primary matter, called hule. On this,
different forms could be impressed, much as a sculptor can make
different statues from one block of marble. These forms can be
removed, and replaced by new ones, so that the idea of the trans-
mutation of the elements arose. Aristotle's elements are therefore
really fundamental properties of matter, and as the most funda-
mental he chose hotness, coldness, moistness, and dryness. By
combining these in pairs, as shown in the diagram, he obtained
what are called the four elements, fire, air, earth, and water :

Moist

Hot

Air

Water

Fire

Earth

Cold

Dry

Thus, water is the type of moist and cold things ; fire of hot and
dry, and so on. To the four material elements a fifth, immaterial,
one was added, which appears in later writings as the quintessence.
This corresponds with the modern ether.

Early alchemy. — The science of Chemistry arose from two
sources :

1. The speculative philosophy of the Greeks.

2. The Egyptian art of working in metals.

The name Chemistry occurs later, and is supposed to be derived
from the word chemi, meaning " black or burnt," or " Egyptian,"
or both.

The Egyptian technique, handed down from the workshops, was
first described in Greek, and afterwards translated into Latin.
Thus, in the Papyrus of Leyden, discovered at Thebes, and pre-
served in the Museum at Leyden, we find many practical recipes.
This papyrus is written in Greek, at a date not accurately known
but supposed to be the third century, and appears to have included
the working notes of a fraudulent goldsmith. Recipes for plating
base metals with gold occur in it, but the author is quite aware
that no real transmutation had occurred. Thus, he says :

" One powders up gold and lead into a powder as fine as flour, 2 parts
of lead for 1 of gold, and having mixed them, works them up with gum.
One covers a copper ring with the mixture ; then heats. One repeats
several times until the object has taken the colour. It is difficult to

ii ELEMENTS, COMPOUNDS, AND SOLUTIONS 29

detect the fraud, since the touchstone gives the mark of true gold. The
heat consumes the lead but not the gold."

In the course of translation of such documents, the language
became confused, and the idea of a real transmutation crept in.
On the conquest of Asia, Africa, and part of Europe, by the Arabs,
the latter assimilated the knowledge of the subject races, and the
study of chemistry was called Alchemy, the prefix al being the
definite article in Arabic. Further translations were made, and
additional errors arose.

Geber. — Geber was an Arabian alchemist living in the ninth
century, but the Latin * writings usually attributed to him
belong to a much later date. The " Latin Geber " added to the
four elements of Aristotle the alchemical elements, sulphur, and mer-
cury ; a third, salt, was introduced by another alchemist called
Basil Valentine, supposed to have written in 1470, but probably
mythical, the real author being a German, Tholde, living in
the seventeenth century, f Sulphur was the principle of com-
bustibility ; salt the fixed part left after calcination ; whilst mer-
cury was the principle of metallicity, contained in all metals.

Gold and silver, according to the Latin Geber, contain a pure
mercury, united with a " clean sulphur," which is red in the gold and
white in the silver. Other metals contain an " unclean sulphur,"
but it was supposed that the base metals could be converted into
gold and silver by altering the proportions of mercury and sulphur
in them and " cleansing " the latter. This process was to be
brought about by a substance called the philosopher's stone, which
was described as a red powder. Some of the recipes for its pre-
paration, in so far as they are intelligible, show that it was an
amalgam of gold, or a solution of gold in mercury, the latter
being driven off in the fire, leaving the gold.

latroehemistry. — In the sixteenth and seventeenth centuries
another school of chemists arose, called the latrochemists, i.e.,
the medical chemists, who attempted to prepare the elixir of lite, which
should cure all diseases, and confer perpetual youth. Paracelsus
(1493-1541) was the founder of this sect; he believed in the
philosopher's stone and the elixir of life. It was thought that the
philosopher's stone and the elixir of life would, when prepared,
turn out to be the same, an idea which no doubt arose partly from
the Oriental imagery of the Arabian alchemists, who spoke of
" healing " metals when they were transmuted, and partly because
many substances, such as arsenic, mercury, and zinc, change the

* ''Liber Geber," British Museum, 1473 (?). English translation : "The
Works of Geber, the Most Famous Arabian Prince and Philosopher,"
Richard Russel, London, 1678.

f Triumph Wagen antimonii. F, Thoelde, Leipzig, 1604.

30 INORGANIC CHEMISTRY CHAP.

colours and properties of metals and also have a powerful action
on the human body.

Experiments on the supposed transmutation included the roasting
of the sub-metallic mineral galena in air, when lead was formed,
with a strong smell of sulphur ; and the production of a small
button of silver when the lead was burnt off by heating on a cupel,
or dish made of bone-ash. Also, if iron pyrites, a yellow mineral
looking somewhat like gold, was melted with lead, and the lead
cupelled, a minute amount of gold was left. Both the silver and
gold, of course, pre-existed in the minerals, and are prepared from
them at the present day. Again, a steel knife-blade dipped into
a solution of blue vitriol (copper sulphate) apparently became
converted into copper.

The later history of Alchemy, however, is mainly that of fraud
practised by the " adepts " on credulous dupes, so that the " science "
ended as it began. One method of effecting transmutation was to
stir the materials in the crucible with a hollow iron rod filled with
gold powder, and stopped with wax.

Attempts at transmutation have been made in quite recent times,
the philosopher's stone in this case being radium. Ramsay and
Cameron (1907) thought they had converted copper into lithium
to a minute extent by exposing a solution of copper sulphate to the
emanation of radium, but Mme. Curie showed that the lithium
came from the quartz vessels used.

Van Helmont (1577-1644) represents the transition from alchemy
to modern chemistry. His writings * show the beginnings of
scientific method, although he still believed in transmutation,
having seen the operation performed once by an adept, and sought
for the alkahest, or universal solvent. He considered that
all materials were derived from water, as taught by Thales
(B.C. 600), and describes an experiment in which a small willow
twig was grown in a weighed pot of earth, supplied only with
water. After five years the tree was weighed, and had gained
164 Ib. in weight, whereas the earth had lost practically nothing.
Hence he concluded that the tree had been formed solely from
water.

It is something of an irony of fate that this erroneous conclusion,
in which the assimilation of carbon dioxide from the air by the
plant was ignored, should have been reached by the discoverer of
that gas. Van Helmont invented the name gas, derived from chaos,
describing the supposed wild motion of its particles, and designated
carbon dioxide as gas sylvestre, i.e., the " gas of the woods," or the
" wild, untamable gas," because, having corked up limestone and
acid in a bottle, he found that the latter was burst by the gas

* " Ortus Medicinae," Amsterdam, 1648 ; Leyden, 1656.

ii ELEMENTS, COMPOUNDS, AND SOLUTIONS 31

sylvestre. A gas, according to Van Helmont, is something which
cannot be kept in a vessel.* In his treatise " deFlatibus " he men-
tions another gas, gas pingue, which is inflammable, and is produced
in fermentation. It was probably impure hydrogen.

Robert Boyle. — Modern chemistry may be said to have begun
with Robert Boyle (1627-1691), and for two reasons. In the first
place Boyle was the first to study chemistry for its own sake, and

ROBERT BOYLE.

not as a means of making gold or medicines. In the second place,
he introduced a rigorous experimental method into chemistry, and
in particular overthrew the doctrines of the Aristotelian and
Alchemical elements, by showing that none of them could by any
process be extracted from metals. In the case of gold, neither

* " Hunc spiritum incognitum hactenus, novo nomine gas voco, qui nee
vasis cogi, nee in corpus visibile reduci potest."

32 INORGANIC CHEMISTRY CHAP.

water nor solvents can extract sulphur or mercury from it : " the
metal may be added to, and so brought into solution or crystalline
compounds, but the gold particles are present all the time, and the
metal may be reduced to the same weight of yellow, malleable,"
ponderous substance as it was before the experiment." Boyle's
definition of an element has already been given (p. 27).

The chemical elements. — The list of substances at present
accepted as elements, which is given on p. 145, comprises eighty-
six names. Of these only about one-half are those of substances
commonly found in chemical laboratories, and of these only about
twenty occur in the uncombined state. About 99 per cent, of
terrestrial matter is composed of some twenty elements and their
compounds.

An estimate of the occurrence of the elements in the air, the sea
and other waters, and the crust of the earth to a depth of twenty •-
four miles, has been made by F. W. Clarke. The following table
gives the average composition by weight of these materials, taken
together, in parts per 100 : —

Oxygen 49-85 Calcium 3-18 Hydrogen 0-97

Silicon 26-03 Sodium 2-33 Titanium 0-41

Aluminium 7-28 Potassium 2-33 Chlorine 0-20

Iron 4-12 Magnesium 2-11 Carbon 0-19

Oxygen is seen to be the most abundant element, accounting
for one-half the* total mass ; silicon, which occurs mainly in the
form of the oxide silica as quartz and sand, and in combination in
many rocks, is the next in abundance. Nitrogen, occurring in the
atmosphere, and the other elements, many of them constituting
living matter, together equal only about 1 per cent, of the whole.

The composition of the centre of the earth is not accessible to
experiment, but since the mean density of the earth is about 5-6,
the central part must consist largely of substances of high density,
one of which is probably iron.

Some of the elements are widely distributed in nature, some in
large quantities, such as oxygen, silicon, sodium, and iron, and
others in very much smaller amounts, such as lithium, rubidium,
and helium. Other elements, such as erbium, occur only in very
small amounts in particular localities.

By means of spectrum analysis (Chap. XXXVI), it is possible to dis-
cover the elements present in the sun and stars. The following elements
have been recognised in the atmosphere of the sun : "aluminium,
barium, beryllium, cadmium, calcium, carbon, cerium, chromium,
cobalt, copper, erbium, germanium, helium, hydrogen, iron, lan-
thanum, lead, magnesium, manganese, molybdenum, neodymium,
nickel, niobium, oxygen, palladium, rhodium, scandium, silicon,

ii ELEMENTS, COMPOUNDS, AND SOLUTIONS 33

silver, sodium, strontium, tin, titanium, vanadium, yttrium, zinc,
zirconium, and nitrogen as cyanogen. The following are doubtful :
iridium, lithium, osmium, platinum, potassium, ruthenium, tan-
talum, thorium, tungsten, and uranium.

The spectra of stars show that these may be divided into groups.

Some stars show dark lines on a bright spectrum ground : others
show bright lines on a faint spectrum background. Great differ-
ences are found in the stellar spectra, and the classification usually
adopted by astronomers is as follows :

Class O (Wolf-Rayet type) : bright lines on a faint continuous back-
ground.

Class B (Orion type) : dark lines of helium sparsely set on a bright
ground.

Class A (Sirian type) : hydrogen lines most conspicuous.

Class F (Calcium type) : hydrogen lines still conspicuous, but many
faint lines of metals appear, notably two strong calcium lines in the
violet.

Class G (Solar type) : numerous strong metallic lines appear, as in
sunlight.

Class K (Sun-spot type): lines darker, and flutings occur, as in sun-
spots. Hydrogen lines faint.

Class M (Fluted type) : flutings due to titanium oxide marked, as well
as flutings due to carbon.

The nebulae show the presence of hydrogen, helium, and possibly
an element, nebulium, not known on the earth.

Lockyer observed that the hotter stars contain fewer elements
than the cooler stars, and he assumed that some of the terrestrial
elements are decomposed at the very high temperatures in the hot
stars into simpler elements, some of which may be the ordinary
elements known to us.

Specimens of extra-terrestrial elements come to us occasionally
in the form of meteorites, which are masses consisting chiefly of
metallic iron, together with nickel, phosphorus, carbon, oxygen,
calcium, silicon, and hydrogen. No new elements are found in
them.

On the whole, therefore, we may assume that the composition
of the sun and stars is similar to that of the earth, or still simpler.

SUMMARY OF CHAPTER II

All parts of a homogeneous pure substance exhibit the same proper-
ties, and behave in the same way, under the same conditions. Pure
substances may become changed into other pure substances, with
different properties. This is the result of chemical change. These

34 INORGANIC CHEMISTRY CH. n

changes may be proved experimentally to depend on the combination
of forms of matter previously distinct, or the separation of distinct
substances from a previous condition of union, i.e., to decomposition.
Pure substances, after having undergone chemical change, may be
recovered, qualitatively and quantitatively the same as they were at
first, by a reverse process of change. This is a result of the Law of
Conservation of Matter, or the Law of Indestructibility of Matter.

Certain substances have resisted all attempts to decompose them,
and in the present state of our knowledge are regarded as chemical
elements, or the simplest distinct forms of matter.

(See Mallet, " Memorial Lecture on Stas," Chemical Society's
Memorial Lectures, 1893.)

EXERCISES ON CHAPTER II

1. Define : compound, element, solution, analysis, synthesis, chemi-
cal change. In what ways does a chemical change differ from a physical
change ?

2. State the Law of Conservation of Matter, and describe two simple
experiments to illustrate its application to chemical changes. To what
degree of accuracy is it known to be true, and how has this been tested ?

3. Trace briefly the evolution of the conception of the chemical
elements. What is known as to the distribution of the elements in the
earth and stars ?

4. On what theoretical and experimental bases was Alchemy founded,
and why has its pursuit been abandoned by chemists ?

5. Why are common salt and water said to be compounds, but the
liquid formed by mixing them together a solution ?

CHAPTER III

THE COMPOSITION OF THE AIR AND THE THEORY OF COMBUSTION

The discovery of gases. — Reference has been made to the two
gases described by Van Helmont (1577-1644), viz., gas sylvestre
(carbon dioxide) and gas pingue (hydrogen). No new gases were
discovered from then until the time of Priestiey (1772), although
the two gases of Van Helmont were carefully investigated by Henry
Cavendish (1766) ; gas sylvestre was named fixed air by Joseph
Black, 1755, and gas pingue inflammable air by Cavendish, re-
spectively. Inflammable air was obtained by the action of
sulphuric and hydrochloric acids on zinc, iron, and tin.
Cavendish observed that the inflammable air was " the
same, and of the same amount, whichever acid is used to
dissolve the same weight of either metal " [iron or zinc], and hence
concluded that the gas came from the metal. He found that
inflammable air was much lighter than common air, whilst carbon
dioxide was heavier. (" On Factitious Airs," Phil. Trans., 1766.)

Joseph Priestley (1733-1804), whose discoveries are recorded in
his " Observations on Different Kinds of Air," * recognised several
new gases. At that time gases were called " airs," Van Helmont's
name, gas, having dropped out of use. Priestley prepared and
examined oxygen, nitrous oxide, nitric oxide, nitrogen dioxide,
hydrochloric acid gas, ammonia gas, and sulphur dioxide. He
improved the familiar pneumatic trough, and was able to collect
over mercury many gases which are very soluble in water (e.g.,
ammonia, and sulphur dioxide).

Priestley's work firmly established the fact that a number of
different gaseous forms of matter exist, each with definite pro-
perties, so that the old idea that such of these as had been
noticed were merely common air mixed with impurities, was finally
abandoned.

Combustion and the calcination of metals. — There are two kinds
of chemical change which, since they were investigated side by side,

* 6 vols., 1774-86 ; abridged edition, 3 vols., 1779-86.

35 D 2

36 INORGANIC CHEMISTRY CHAP.

and depend on the same cause, may conveniently be described
together. These are combustion, and the calcination of metals.

The alchemists attached great importance to the effects of heat
on substances, and their writings describe many types of furnaces,
and experiments made with them. The metals, except gold and
silver, were found to change when heated in open crucibles, and to
leave a dross, which was called a calx (Latin calx, lime). It was
noticed in the sixteenth century that this calx is heavier than the
metal : the explanation usually given was that fire, or caloric,
possessed weight, and was absorbed by the metal in forming the
calx. Jean Key (1630) " devoted several hours to the question,"
without apparently making any experiments, and concluded that
the air becomes thickened or adhesive by the action of the fire,
and sticks to the metal. His ideas are very crude and inaccurate.

Nitre air. — Robert Boyle * (1673) heated tin in a glass retort, and
when it was melted, sealed off the neck and continued the heating
for two hours. The retort was cooled, and the sealed tip of the
neck broken. Air rushed in, " because when the retort was sealed,
the air within it was highly rarefied." Boyle, from his method of
experimenting, therefore did not notice, as Lavoisier did a century
later, that some of the air was absorbed, and that the tin had
increased in weight.

Boyle then showed that when sulphur was sprinkled on a red-hot
plate under an exhausted air-pump receiver, it smoked but did not
burn. On admitting air, " divers little flashes were seen." But if
gunpowder were sprinkled on the hot plate under the vacuous
receiver, he saw " a pretty broad blue flame, like that of
brimstone, which lasted so long as we could not but wonder
at it." Gunpowder could also burn under water. Boyle,
therefore, somewhat reluctantly, concluded that a flame can
exist without air, and that the increase in weight of metals
on calcination is due to their absorption of caloric, or fire,
which he considered to be material, and capable of being
weighed in a balance. He observed that if charcoal is strongly
heated in a closed retort it does not burn, but the caput mortuum
(a fanciful name due to the alchemists, who represented a residue
by the symbol of the skull and crossbones) becomes black again on
cooling. If, however, air is admitted, the charcoal burns, and
crumbles down to white ashes.

The latter experiment was repeated by Robert Hooke (at one
time an assistant to Boyle), who, in his " Micrographia " (1665), put
forward the first rational theory of combustion. Hooke found that
a bit of charcoal or sulphur burns brilliantly when thrown into fused
nitre.

* Works, edited by Birch, 5 vols., 1744; abridged by Boulton 4 vols.,
1699-1700 ; do. by Shaw, 3 vols., 1738.

in COMPOSITION OF AIR— THEORY OF COMBUSTION 37

EXPT. 22. — Fuse about 5 gm. of nitre in each of two test-tubes,
supported by clamps over a tray of sand. Throw into one a small piece
of charcoal ; this swims about and burns brightly. Into the other
throw a' small piece of sulphur : this burns with a beautiful blue flame.

On the basis of these experiments Hooke founded his theory of
combustion, which was briefly as follows :

" (1) Air is the universal dissolvent of all sulph~a3X)US [i.e., com-
bustible] bodies. (2) This action of dissolution produces a very
great heat, and that which we calibre. (3) This dissolution is made
by a substance inherent and mixed with the air that is like, if not
the very same with, that which is fixed in saltpetre [nitre]." This
substance he called nitre air. In this way he was able to explain
the combustion of gunpowder, one constituent of which is nitre,
in the absence of air.

John Mayow (" Tractatus quinque medico-physici," Oxford
1674) elaborated a theory similar to that of Hooke, but supported by
descriptions of experiments (which were not published by Hooke).
He concluded that air consists of two gases ; one is the nitre-air
of Hooke, called by Mayow spiritus nitro-aereus, which is concerned
in combustion and respiration ; and the other is an air incapable
of supporting either of the latter.

The experimental evidence was as follows : —

(1) EXPT. 23. — Mayow inverted a large glass globe over a lighted
candle standing in water, equalising the levels of the latter by means
of a siphon, which was then quickly withdrawn. The water rose inside
the globe, showing that some air had disappeared. When the candle
was extinguished, a large bulk of air was left, but this would not
support the combustion of sulphur or camphor lying on a small
shelf inside the globe, when they were heated by a burning glass.
(Fig. 26.)

(2) A mouse when introduced into the residual gas died.
Conversely, when a lighted candle was plunged into a confined volume
of air in which a mouse had died it was instantly extinguished. If a
mouse was kept in a vessel of air closed by a bladder ("Fig. 27), the con-
traction of the air was perceptible.

(3) Gunpowder rammed into a paper tube and ignited continued to
burn under water. The air fixed in nitre can therefore take the place
of ordinary air in supporting combustion, and since things burn more
brilliantly in fused nitre than in common air, the nitre must contain an
abundant supply of nitre air, which is the part of common air concerned
in combustion.

(4) Mayow repeated an old experiment described in Libavius'
" Alchymia'' (1595), viz., calcining a cone of metallic antimony on a

38 INORGANIC CHEMISTRY CHAP.

marble slab by means of a burning-glass. Although abundant fumes
were evolved, the calx weighed more than the metal. The calx was
found to be identical with that formed by the action of nitric acid on the
metal.

Mayow did not succeed in isolating nitre air, and although Hooke,
in his " Lampas " (1677), says that his theory was generally received
(a similar theory was in fact mentioned by Lemery in his " Cours
de Chemie," 1675), these beginnings of a true theory of combus-
tion were soon stifled by an erroneous dogma, due to two German
chemists, which persisted for a century, and obscured nearly
every branch of chemical science. This was the famous theory
of phlogiston, of Becher and Stahl.

Theory of phlogiston. — It was a favourite expression of the

FIG. 26.— Mayow's Experiment FIG. 27.— Mayow's Experiment

on Combustion. on Respiration.

alchemists that inflammable bodies contain sulphur : ubi ignis et
color ibi sulphur. John Joachim Becher, in his " Physicse sub-
terranese " (1669), remarked that the constituents of bodies are
air, water, and three earths, one of which is inflammable (terra
pinguis) ; the second mercurial ; the third fusible, or vitreous.
These correspond with the sulphur, mercury, and salt of the
alchemists. On combustion, the " fatty earth " burns away.

In 1702 Becher s treatise was republished, with a long intro-
duction, by George Ernst Stahl, professor at Halle. Stahl was
a good chemist and an excellent teacher, and in his lectures and
text-book (" Fundamenta chymiae," 1723), he popularised Becher 's
views in an improved form. He gave the name phlogiston (from
the Greek <£Ao£ = flame) to the terra pinguis of Becher. When
bodies burn^ or are calcined, phlogiston escapes with a rapid

Ill

COMPOSITION OF AIR— THEORY OF COMBUSTION

whirling motion ; when the original bodies are recovered by
reduction, phlogiston must be replaced. Oil, wax, charcoal, and
sulphur are all rich in phlogiston, and may be used to restore it
to a burnt material. Zinc on heating to redness burns with a
brilliant flame, hence phlogiston (<£) escapes. The white residue is
calx of zinc. If it is heated to whiteness with charcoal (rich in
phlogiston) zinc distils off. Hence : calx of zinc -j- <f> = zinc.
Similarly with other metals. If phosphorus is burnt, it produces
an acid matter, and much heat and light are evolved. Hence :
phosphorus = acid -f- <£• If the acid is heated with charcoal,
phlogiston is absorbed and phosphorus is reproduced.

Stahl's theory united a great many previously isolated facts, and
became almost universally accepted during the eighteenth century.

During this

period the in- _ d

crease in weight
of metals on
calcination was
usually ignored
as of little im-
portance, or as
belonging to
physics rather
than to chemis-
try, although the
fact was destined
later to overturn
the whole theory
of phlogiston.

EXPT. 24.— This
increase of weight
is readily shown
by the following
is taken up by

FIG. 28.— Increase in Weight of Iron on Burning.

experiment. Some finely divided reduced iron
a horse-shoe magnet counterpoised from one
arm of a sensitive balance (Fig. 28), a piece of asbestos paper being
placed in the pan underneath the magnet. If a spirit-lamp flame
is applied to the tufts of iron adhering to the magnet, the powder
begins to glow, and after calcination falls from the magnet. The pan
on the side of the balance where the magnet is suspended sinks,
showing that the iron increases in weight during calcination. The
iron calx left is found to be black in colour, whereas the original iron
powder is grey.

Scheele's experiments on fire and air. — Carl Wilhelm Scheele
( 1742-1786) was a firm believer in Stahl's theory. A poor apothecary

40 INORGANIC CHEMISTRY CHAP.

of Stockholm, he made a great number of chemical discoveries of
the very first rank, those on combustion being published in his
treatise " On Air and Fire." These experiments were made
chiefly before the autumn of 1770, and all prior to 1773. The MS.
reached the printers in 1775, but owing to delay the book did not
appear until 1777, when many of Scheele's discoveries had been
made independently, and published, by Priestley in England.
Scheele's priority was only established in 1892, from his original
laboratory notes, discovered at Stockholm.

In his first set of experiments Scheele noticed the contraction of a
confined volume of air standing in contact with various materials.
He used, for instance, a solution of liver of sulphur (hepar sulphuris),
a solution of sulphur in lime-water, linseed oil, and iron filings
moistened with water, all of which, he observes, are rich in
phlogiston, or, as he called it, the inflammable substance. In all
cases there was a loss of air.

A solution of sulphur in potash, which is yellow, became colourless
in contact with air, and the solution contained " vitriolated tartar,"
which could be formed from potash and sulphuric acid. No sulphur
was left over.

EXPT. 25. — Take three glass tubes, 2 ft. long and f in. wide, fitted
with rubber stoppers, 'and divided into five equal volumes by strips of
label. In one place a moistened piece of liver of sulphur (made by fusing
potassium carbonate with flowers of sulphur in a covered crucible till
evolution of gas ceases), and in the second a piece of phosphorus stuck
on a piece of copper wire. Wet the inside of the third tube with
water, and sprinkle it with clean iron filings. Cork the three tubes
and allow them to stand inverted in three large glass cylinders of
water for a few days (Fig. 29). Open the tubes under water, and
observe that the latter rises in the tubes until one -fifth of the volume
is occupied. Cork the tubes, remove them from the cylinders, and
insert a lighted taper into the gas in each. The flame is extinguished.

The inflammable substance was not contained in the residual gas,
which differed from common air. For, if this gas had been formed by
the union of common air with phlogiston, and contraction, it should
be denser than common air. But : "a very thin flask which was
filled with this air, and most accurately weighed, not only did not
counterpoise an equal volume of ordinary air, but was even some-
what lighter." Thus. " the air is composed of two fluids, differing from
each other, the one of which does not manifest in the least the
property of attracting phlogiston, whilst the other, which composes
between the third and fourth part of the whole mass of the air, is
peculiarly disposed to such attraction." These two constituents
of common air Scheele called Foul Air, and Fire Air, respectively.

Ill

COMPOSITION OF AIR— THEORY OF COMBUSTION

Scheele next placed a little phosphorus in a thin flask, corked the
latter, and warmed it until the phosphorus took fire. A white
cloud was produced, which attached itself to the sides of the flask
in white flowers of " dry acid of phosphorus." On opening the
flask under water, the latter rushed in, and occupied a little less
than one-fifth of the flask [ExpT. 26]. By allowing phosphorus
to stand for six weeks in the same flask, until it no longer glowed,
about one-third of the air was lost.

Scheele then burnt a hydrogen flame under a glass globe standing
over water (Fig. 30). The water at once began to rise, until it
filled one-fourth of the flask, when the flame went out.

FIG. 29. — Di munition of
Air by Phosphorus.

FiG. 30. — Scheele's Experiment

on the Combustion of

Inflammable Air.

EXPT. 27. — Burn a jet of hydrogen from a Kipp's apparatus (p. 185)
inside a graduated bell -jar over water. The gas is turned off as soon
as the flame goes out, and, after cooling, it will be found that one-fifth
of the air has disappeared (Fig. 31).

Scheele thought that hydrogen (inflammable air) was phlogiston,
and in considering the last experiment he asked himself :

(1) What has become of the fire air ?

(2) Where has the phlogiston (inflammable air) gone ?

The fire air, he argued, must either remain in the air, be dissolved
in the water, or have escaped through the vessel. (He did not notice
the moisture condensed on the flask, which contained both the
missing gases, because he worked over hot water, which itself gave
off steam.) The residual foul air was lighter than common air,

INORGANIC CHEMISTRY

CHAP.

although the latter had undergone a contraction, hence the two
substances cannot be present in it. Further, he found nothing
in the water. Scheele therefore concluded that the fire air and
phlogiston had escaped through the glass, combined in the form of
heat and light, which he considered to be material and called
caloric: fire (or caloric) = fire air -f <£.

This hypothesis, of course, is quite incorrect, but it led Scheele to
the most important discovery that has ever fallen to the lot of a
chemist, viz., the isolation of "fire air." It is by no means un-
common to find an important discovery resulting directly from a
false assumption.

Scheele now set himself to reverse the change he thought had

FIG. 31. — Combustion of Hydrogen in Air.

taken place, i.e., to decompose caloric (or heat) into fire air and
phlogiston. For this purpose it was necessary to present to the caloric
a substance having a greater attraction for phlogiston than is
exhibited by fire air. The latter should then be set free. For this
substance he chose nitric acid, because it readily corrodes metals,
taking out their phlogiston, and forming red fumes. In order to
subject it to the action of- caloric, the acid must be fixed, and
Scheele did this by combining it with potash. In order to set the
acid free again at the high temperature, he distilled the resulting
nitre with strong oil of vitriol (sulphuric acid) in a retort (Fig. 32).
Brown fumes came off, which were absorbed in a bladder containing
milk of lime, attached to the neck of the retort. The bladder gradu-

COMPOSITION OF AIR— THEORY OF COMBUSTION

ally filled with a colourless gas, in which a taper burned with a
flame of dazzling brilliance. This was fire air — the " nitre-air "
which had eluded Hooke and Mayow.

Scheele prepared fire air in a variety of other ways. Thus, he heated
calx of mercury (mercurius calcinatus per se), which he supposed absorbed
phlogiston from the caloric, setting free the fire air :

Calx of Mercury +• (<ft + Fire Air)
Caloric

-f Calx of Mercury) -f Fire Air.

Metallic Mercury.

He also obtained fire air by heating :

(1) Black manganese (manganese dioxide) with sulphuric or arsenic
acid [ExpT. 28].

(2) Nitre- alone
strongly. This
gives fire air, and
a residue evolving
red fumes with
acid [ExpT. 29].

(3) Silver or
mercurous car-
bonate, the aerial
acid (carbon di-
oxide) simultane-
ously produced

FIG. 32.— Scheele's Experiment : Isolation of Fire Air.

being absorbed
by means of an
alkali : silver carbonate = silver + fire air + aerial acid.
(4) Magnesium and mercurous nitrates [EXPT. 30].

Scheele found that fire air is completely absorbed by moist liver
of sulphur. When he burnt phosphorus in a thin flask of it, the
flask burst on cooling. With a thicker flask, the cork could not be
taken out under water, but could be pushed in, when water rushed
in and filled the flask. A hydrogen flame continued burning in
the gas until seven-eighths were absorbed.

When fire air was added to the foul air left after combustion of
hydrogen, etc., in air, so as to restore the original volume, the
mixture had all the properties of ordinary air, e.g., it left the same
residue after standing over liver of sulphur.

EXPT. 31. — Fill a gas-jar, divided into 5 parts, four-fifths with nitro-
gen from a gas-holder, and then fill Up the remaining one-fifth with
oxygen. Test the gases separately in tubes with a taper, and then the
mixture.

INORGANIC CHEMISTRY

CHAP.

Scheele placed various animals and insects in confined volu mes of
air, taking care to put along with them their appropriate foods.
He found that they ultimately died ; aerial acid (Black's fixed
air] was produced, and a contraction of the air resulted, the residue
extinguishing a flame. Similar results were found with sprouting
peas. Two large bees were placed in a bottle of fire air over milk
of lime, Scheele having " provided some honey for their stay."

JOSEPH PRIESTLEY.

After eight days the bottle was almost completely filled with
liquid, and the bees were dead. He also noticed that the fire air
is partly dissolved out of common air when this stands over
water which had been boiled. A candle burns more brightly in the
air expelled from the water by boiling than in common air.

Priestley's experiments on dephlogisticated air. — Priestley, having
come into the possession of a powerful convex lens, or burning-glass,
tried by its aid to extract " air " from various chemicals given

Ill

COMPOSITION OF AIR— THEORY OF COMBUSTION

to him by his friend Warltire. Among these was red precipitate, or
mercurius cakinatus per se, obtained by heating mercury in air, the
nature of which had long been a puzzle to chemists. The sub-
stances were heated by focussing the sun's rays on them in small
phials (Fig. 33) filled with, and inverted over, mercury.

" Having procured a lens of twelve inches diameter, and twenty
inches focal distance [the statue of Priestley at Birmingham, in which
he is represented as performing his
famous experiment, shows, in error, a
very much smaller lens than this], I
proceeded with great alacrity to ex-
amine, by the help of it, what kind of
air a great variety of substances, natural
and factitious [i.e., artificially prepared :
cf. Cavendish's factitious airs] would
yield . . . With this apparatus, after
a variety of other experiments, . . .
on the 1st August, 1774, 1 endeavoured
to extract air from'mercurius calcinatus
per se ; and I presently found that, by
means of this lens, air was expelled from
it very readily. Having got about
three or four times as much as the
bulk of my materials, I admitted

water to it, and found that it was not imbibed by it. But what sur-
prised me more than I can well express, was, that a candle burned in
this air with a remarkably vigorous flame ... I was utterly at a loss
how to account for it."

Priestley's haphazard method of work is clear from this quotation :
in another place he remarks that in his discoveries "more is owing
to what we call chance, that is, philosophically speaking, to the observa-
tion of events arising from unknown causes, than to any proper design
or preconceived theory in this business."

Priestley found that a mouse lived twice as long in the new air
as in the same confined volume of common air, and revived after-
wards when taken out. He breathed it himself, and fancied his
" breast felt peculiarly light and easy for some time afterwards "
— hence he recommended its use in medicine (it is now used in the
treatment of gas poisoning and pneumonia).

Priestley, who was a minister of religion, was doubtful whether we
might not " live out too fast " in it, and remarks that : " The air which
nature has provided for us is as good as we deserve." He suggested
that by blowing fires with the new air, very high temperatures might

FIG. 33. — Isolation of Oxygen by
Priestley (1774).

46 INORGANIC CHEMISTRY CHAP.

be attained, and his friend Michell was later on able to melt platinum
in this way.

Priestley now asked himself : " What is this new air ? " He
assumed, from the teachings of Stahl, that a candle on burning
gives out phlogiston, and is extinguished in a closed vessel after a
time because the air becomes saturated with phlogiston. Ordinary
air, therefore, supports combustion because it is only partially
saturated with phlogiston, and can absorb more of it. Substances
burn in air with only a moderate flame, whereas in the new air
the flame is vivid ; Priestley, therefore, concluded that the new
gas must contain little or no phlogiston, and hence he called it
dephlogisticated air. The gas left when bodies burnt out in ordinary
air was named, for a similar reason, phlogisticated air :

Phlogisticated Air [Nitrogen] = Air + 0. (Scheele's Foul Air.)
Dephlogisticated Air [Oxygen] = Air — <f>. (Scheele's Fire Air.)

Priestley believed that " phlogiston is the same thing as in-
flammable air, and is contained in a combined state in metals,
just as fixed air is contained in chalk and other calcareous sub-
stances ; both being equally capable of being expelled again in
the form of air [by the action of acids]."

Lavoisier and the Antiphlogistic Theory.— -Antoine Laurent
Lavoisier (1743-1794), the famous French scientist, began his
experiments on combustion in 1772. He found that when sulphur
and phosphorus are burnt in closed glass tubes they do so at the
expense of part of the air, since

(a) if the tube be afterwards opened under water, the latter
rushes in and partially fills the vessel ;

(6) if opened in the air, the latter rushes in, and the vessel
becomes heavier.

He concluded that both these substances on burning take something
irom the air.

Lavoisier next modified Boyle's experiment of calcining tin and
lead, by using weighed sealed retorts. He ob tamed the same results
as with sulphur and phosphorus, and drew the same conclusion.
On heating the calx of lead with charcoal he found it lost in weight,
and " an air was abundantly evolved." Thus something is taken
from the calx in forming the metal, and this must be "an air."
Further Lavoisier could not go.

But in October, 1774, Priestley visited Paris with Lord Shelburne,
and told Lavoisier at dinner of his discovery of dephlogisticated
air, saying he " had gotten it from precip. per se and also red lead " ;
whereupon, he says, " all the company, and Mr. and Mrs. Lavoisier
as much as any, expressed great surprise." In Lavoisier's note-
book of 1775 there occurs an entry dated 13th February, recording

in COMPOSITION OF AIR— THEORY OF COMBUSTION 47

an experiment on " precipite per se de chez M. Baume," and men-
tioning the disengaged gas as " Fair dephlogistique de M. Prisley "
(sic). When, therefore, Lavoisier, in his ' Traite de Chemie "
(1789), speaks of " this air, which Dr. Priestley, Mr. Scheele, and
I discovered about the same time," one is compelled to dissent.
There is no evidence that Lavoisier had any claim to be regarded
as a discoverer of oxygen gas.

LAVOISIER.

Lavoisier was quick to see the important bearing of Priestley's
discovery on his own unfinished work ; he was able to prove that
it is dephlogisticated air which is absorbed in the calcination of
metals, by a famous experiment, described in his "Traite"
(1789).

He heated 4 oz. of mercury in a retort which communicated
with a measured volume of air in a bell- jar over mercury (Fig. 34).
The volume of air in the bell and in the retort was 50 cu. in. After

48 INORGANIC CHEMISTRY CHAP.

a time he noticed the formation of red specks, and scales, of calx
on the surface of the mercury. After twelve days the scales no
longer increased ; the fire was removed, and the experiment stopped.
The air had contracted to 42 cu. in., and the gas left was " mephitic
air," which Lavoisier at first called atmospheric mofette. The
scales, or mercury calx (mercurius calcinatus per se), were collected
and found to weigh 45 grains. They were transferred to a small
retort and heated ; 8 cu. in. of dephlogisticated air, which was
" an elastic fluid, much more capable of supporting respiration
and co'mbustion than ordinary air," and hence called by Lavoisier
vital air, or air eminently respirable, were obtained, together with
41 J grains of mercury. When this vital air was added to the atmo-
spheric mofette, ordinary air was formed without any evolution

FIG. 34.— Demonstration of the Composition of Air by Lavoisier (1789).

of heat or light, hence air is probably simply a mixture of these
two gases (as had previously been suggested by Scheele).

Lavoisier made experiments on the combustion of substances
in vital or " pure " air, and summed up his conclusions in the
four statements :

(1) Substances burn only in pure air.

(2) Non-metals, such as sulphur, phosphorus, and carbon, produce
acids on combustion ; hence the gas was called oxygen (o£us = acid).

(3) Metals produce calces on absorption of oxygen.

(4) Combustion is in no case due to an escape of phlogiston, but
to chemical combination of the combustible substance with oxygen.

These statements comprise the fundamental tenets of the anti-
phlogistic theory.

m COMPOSITION OF AIR— THEORY OF COMBUSTION 49

EXPT. 32. — Lavoisier's experiments may be repeated by burning
sulphur, phosphorus, and carbon in jars of oxygen, the substances being
held by deflagrating spoons, and shaking the products with litmus.
The latter is reddened. Magnesium ribbon burns with a blinding light,
giving a white calx, which turns moist red litmus paper blue.

Lavoisier's conclusions were not accepted at once ; Black in
England, and a few French chemists, supported them, but there was
one great difficulty still to be overcome, viz., that the phlogistic
theory could explain a set of experiments which the antiphlogistic
theory could not. A metal like tin or zinc dissolves in an acid giving
inflammable air, and a salt is left on evaporating the solution, which,
on strong heating, parts with its acid and leaves the calx of the
metal. The same salt is formed when the calx is dissolved in the
acid, but no inflammable air is then evolved. Whence comes
the inflammable air in the first experiment ? This was an easy
question for the phlogistonists. " Inflammable air," said they,
" is phlogiston ; the metal is (calx + phlogiston) ; and the salt is
(calx -f acid). In the first experiment you have, clearly :

(calx + $) + acid = (calx + acid) -t- 0

metal salt inflammable air

in the second :

calx + acid = (calx + acid)."

This difficulty was serious : Lavoisier was unable to offer an
explanation. The key was first supplied by the researches of
Cavendish on the formation of water from inflammable air and
dephlogisticated air.

SUMMARY OF CHAPTER IH

The investigation of gases, different from air, made by Henry Caven-
dish in 1766, and Joseph Priestley (1774-86), was of great importance
to chemistry. The theory of combustion and the calcination of metals
due to Robert Hooke and John Mayow, in the seventeenth century,
attributed these changes to the absorption of a gas from the atmosphere,
which, since it is also fixed in nitre, was called nitre air. The theory of
phlogiston, propounded in the next century by Becher and Stahl, ex-
plained the changes as due to the escape from the burning body of a
subtle principle, called phlogiston.

The isolation of nitre air by Scheele (1772), and independently by
Priestley (.1774), enabled Lavoisier to overturn the theory of phlogiston,
and to show that, combustion consists in the union of the combustible
substance with oxygen (nitre air), which is contained in the atmo-
sphere to the extent of one-fifth of its volume.

50 INORGANIC CHEMISTRY OH. in

EXERCISES ON CHAPTER III

1. Describe briefly the experiments of Boyle, Hooke, and Mayow on
combustion, and state their conclusions.

2. Give an account of the theory of phlogiston, and show by an
example how it was applied in the explanation of chemical changes.
What experiments led to the downfall of the theory ?

3. Describe the work of Scheele which led to the isolation of fire-air
(oxygen), and contrast the method used with Priestley's discovery of
dephlogisticated air.

4. Describe the experiment of Lavoisier which proved that common
air contains oxygen and nitrogen. What is meant by the " Anti-
phlogistic theory"?

CHAPTER IV

THE COMPOSITION OF WATER

The work of Cavendish. — Priestley in 1781 observed that when
a mixture of dephlogislicated air (oxygen) and inflammable air
(hydrogen) is ignited it explodes violently. " Warltire noticed that
the sides of the bottle, after the explosion, are bedewed with
moisture.

EXPT. 33.* — Collect a mixture of 2 vols. of hydrogen and 1 vol. of
oxygen in a strong soda-water bottle over water, draining out as much
water as possible from the bottle. Insert a lump of fused calcium
chloride in the bottle arid cork it. When the moisture has been absorbed
by the calcium chloride wrap the bottle in a strong towel, and ignite the
gas by a taper. There is a loud explosion, and the inside of the bottle
becomes filled with steam.

By firing the gases in a copper globe with the electric spark,
Priestley thought he found that there was a slight loss of weight,
which he put down to the escape of caloric (p. 36).

Cavendish in 1781 ignited a mixture of common air and in-
flammable air in a glass globe by means of the spark. He found
that, with 423 vols. of inflammable air to 1000 vols. of common
air, " almost all the inflammable air and about one-fifth part of
the common air, lose their elasticity, and are condensed into the
dew which lines the glass." There was no change in weight
after explosion. He found the ratio of the combining volumes of
hydrogen and oxygen to be 202 : 100.

To examine the nature of the dew, Cavendish performed an
experiment similar to the following.

EXPT. 34. — Burn a jet of hydrogen, dried by calcium chloride, under
a glass retort, cooled by circulating cold water, as shown in Fig. 35.

" This and similar experiments must be performed with adequate precau-
tions to prevent injury in case the bottle should burst. The bottle is wrapped
in a strong towel, with a short length of neck only projecting, and the whole
placed in a strong tin can or iron mortar. A long taper is used.

51 E 2

INORGANIC CHEMISTRY

CHAP.

FIG. 35.— Formation of Water by Combustion
of Hydrogen.

Notice the collection of moisture on the outside of the retort. This
runs down, and may be collected in a small dish. It will be found that
this liquid is odourless, tasteless, boils at 100°, and leaves no residue on
evaporation. It is water.

Cavendish now prepared a mixture of 195 vols. of dephlogisticated

air and 370 vols. of inflammable
air in a bell-jar over water.
The end of a siphon tube, at-
tached to the previously ex-
hausted glass firing-globe or
eudiometer (Fig. 36), was
covered with a bit of wax and
passed inside the jar. The
wax was knocked off, and on
opening the stopcock the globe
was filled with the mixture.
The cock was closed, and the
mixture fired by a spark. The
gas " lost its elasticity," and

on opening the stopcock the globe was again

filled with the gas, which took the place of that

converted into liquid water by the explosion.

This was repeated six times, and water was

produced, which, however, was distinctly acid.

Cavendish proved that the acidity was due to
nitric acid. It was only formed when the oxygen
was in excess, and was due to the combination
with oxygen of nitrogen present in it as an
impurity. By sparking air over water, the latter
was found to contain nitric acid. Acid is not
produced in the explosion of hydrogen with
common air, because the flame is then not hot
enough. If a slight excess of hydrogen is used
with oxygen containing a little nitrogen, no acid
is produced, since it is reduced, if formed, by the
hydrogen.

On account of his attempts to find the cause
of the acidity of the water, Cavendish delayed
publication of his memoir- until 1784. His con-
clusions were curious : "I think we must allow
that dephlogisticated air is in reality nothing but
dephlogisticated water ; . . . and that inflammable air is either
pure phlogiston, as Dr. Priestley and Mr. Kirwan suppose, or else
water united to phlogiston . . . the second of these explanations

seems much the most likely."
formation of water as follows :

THE COMPOSITION OF WATER 53

He therefore represented the

Inflammable air = water -|- <£
Dephlogisticated air— water — <f»

2 water.

CAVENDISH.

The ratio of the combining volumes of hydrogen and oxygen found
in these experiments was 201 : 100.

Cavendish's choice of (water -f <£) for inflammable air was based on
the circumstance that it requires a red heat to start the combination of
the two gases, whereas nitric oxide (cf. p. 578) combines at the ordinary
temperature with dephlogisticated air, and in presence of moisture
forms nitric acid. Nitric oxide, produced by the action of copper on

INORGANIC CHEMISTRY

CHAP.

nitric acid, was regarded as (nitric acid + </>), and it is not likely that
dephlogisticated air should be able to separate <p from its combination
with nitric acid but not able to unite with free $ (if this is inflam-
mable air) under the same conditions. Hence inflammable air is
probably not pure phlogiston, but phlogisticated water,

Cavendish therefore thought that water pre-existed in the two gases,
and its formation on explosion was simply due to a transfer of phlogiston.

James Watt is usually credited with stating, in a letter published
in 1784, that water is composed of the two gases, but Sir E. Thorpe
(Brit. Assoc. Rep., 1890) has given reasons for doubting this.

Lavoisier's explanation of Cavendish's experiments.— n
Lavoisier had been considerably puzzled by the product
of the combustion of hydrogen in oxygen, which he
thought must be an acid. In 1783 he resolved to make
the experiment of burning hydrogen in oxygen on a

FIG. 37. — Decomposition of Steam by Red-hot Iron.

larger scale, so that the product, whatever it was, should not
escape his notice. In May or June of that year, however,
Sir Charles Blagden, who was formerly Cavendish's assistant,
visited Lavoisier, and told him of Cavendish's experiments.
Lavoisier at once saw the importance of the result, and on June
24th, 1783, he repeated the experiments in the presence of Blagden.
On the following day an account of them was sent to the French
Academy of Sciences, and was published in the Memoir es which were
dated 1781 . Practically no mention is made of Cavendish, whose paper
did not appear, for reasons just given, until 1784. Lavoisier's
claims to the discovery of the composition of water were, however,
dismissed by his countryman Arago as pretentious. To Lavoisier,
nevertheless, must be accorded the credit of having first clearly
stated the results. In 1788 he says : " Water is nothing but
oxygenated hydrogen, or the immediate product of the combustion
of oxygen gas with hydrogen gas, deprived of the light and caloric
which disengage during the combustion."

THE COMPOSITION OF WATER

In 1784 Lavoisier and Meusnier decomposed water by passing
its vapour over iron borings heated to redness in a gun-barrel.
Hydrogen was liberated, and the iron converted into the same
black oxide as is produced when iron wire burns in oxygen.

EXPT. 35. — A piece of weldless iron pipe is loosely packed with iron
turnings, and placed in a combustion furnace (Fig. 37). Rubber
stoppers are fitted to the two ends of the pipe (which should project a
fair distance from the furnace so as not to get too hot) and connected
with a flask of water at one end, and an empty flask and gas delivery
tube at the other, as shown. Heat the iron tube to redness and boil the
water in the flask. Water collects in the empty flask, showing that the
decomposition is not complete, but bubbles of gas are evolved from the
delivery tube. Collect a jar of the gas, and show that it is hydrogen.
After the experiment, examine the residue in the tube.

EXPT. 36. — The decomposition of steam by magnesium may be shown
by inserting a piece of
burning magnesium ribbon
into a large conical flask
in which a little water is
boiling vigorously. The
metal burns brightly in the
steam, and the hydrogen
produced burns at the
mouth of the flask ; white
magnesium oxide is left
after the combustion.

Monge in 1783 ex-
ploded hydrogen and
oxygen, drawn from two
graduated jars, in a pre-
viously evacuated glass
globe with firing wires
(Fig. 38). No fewer than
370 successive explosions
were made, producing
four ounces of water, and
the hydrogen and oxygen
combined in the ratio of

by volume. The Tne gases were coiiected through syphon-tubes, pr, PR,
result IS less accurate than in the graduated cylinders, G and H ; they passed through
+Vi of rvF PairoTirlioVi the stopcocks, I and K, to the globe, M, previously ex-

hausted through the tap, L, leading to an air-pump at O.

Lavoisier was now able

to explain the difficulty mentioned on p. 49, and so remove the
last argument against the antiphlogistic theory. A metal, such as

Fia. 38. — MONGE'S EXPERIMENT ON THE
COMBINATION OF HYDROGEN AND OXYGEN GASES.

INORGANIC CHEMISTRY

CHAP.

zinc, when it dissolves in dilute acid, decomposes the water,
liberating hydrogen and combining with oxygen to form the calx
(oxide), which then unites with the acid to form a salt. The
origin of the inflammable gas was therefore cleared up. Lavoisier
regarded the acid as an oxide ; at present it is regarded as (oxide +
water), so that the hydrogen really comes from the acid.

From 1785 the theory of phlogiston gradually disappeared.
At the beginning of the nineteenth century practically every
chemist, except Priestley and Cavendish (whose work had done

so much to overturn it), had
abandoned the theory, and
the science of chemistry as
we know it to-day had its
origin in Lavoisier's writings.
Its foundations had been
laid by the investigations of
Priestley, Cavendish, and
Scheele, but it required the
clear and original mind of
the great French chemist to
form these into a logical and
harmonious system.

The electrolysis of water.—
In 1801 Nicholson and Carlisle,
when experimenting with the
newly-invented electric bat-
tery, discovered that if two
gold wires connected with the
copper and zinc poles of the
battery are dipped into water,
bubbles of oxygen and hydro-
gen, respectively, rise from
these wires. If copper or
iron wires are used, only
hydrogen comes off ; the oxy-
gen is absorbed by the wire,
in 1802 collected the gases

FIG. 39.— Electrolysis of Water.

producing an oxide. Cruickshank

separately, and found that 2 vols. of hydrogen and 1 vol. of oxygen
were liberated. This agrees with Cavendish's result of the synthesis
of water. Davy in 1806 showed that if very pure water is elec-
trolysed in a gold vessel, and the experiment carried out in a
vacuous receiver, so that no impurities can enter from the air, or
be dissolved from glass or other substances of ordinary vessels,
then nothing but hydrogen and oxygen are produced. Thus
water is decomposed by the electric current into hydrogen and oxygen in the
ratio of 2 to 1 by volume.

THE COMPOSITION OF WATER

FIG. 40. — Details of
Platinum Electrode.

EXPT. 37. — An apparatus for the decomposition of water by the

current, or the electrolysis of water, is shown in Fig. 39. It is called a

voltameter or coulometer, and consists of two graduated glass tubes,

with stopcocks above, connected by a horizontal tube, carrying a funnel

for filling the apparatus with dilute sulphuric acid. The electrodes

for leading the current into and out of the liquid consist of pieces of

platinum foil, welded to stout platinum wires

sealed into bent glass tubes inserted through

rubber corks (Fig. 40). These tubes are filled

with mercury, copper wires dipping into which

are connected with binding screws in the

wooden stand. To these binding screws the

wires from the source of current are attached.
Direct current may conveniently be taken

from the supply mains at 110—220 volts, a

lamp -resistance being inserted in the circuit.

If no such current is available, four or six

bichromate cells or accumulators in series will

be found suitable.

Bubbles of gas rise from each electrode ; that

coming from the positive wire, although it

appears more abundant because it is liberated

in smaller bubbles, will be found to occupy very

slightly less than half the volume of the other gas, and, if allowed to

escape from the tap on to a glowing chip of wood, will rekindle the

latter. This gas is oxygen. The other gas, evolved from the negative

wire, when ignited by a taper, burns with a
blue flame, and is hydrogen. Thus, water
is decomposed by electrolysis into 2 vols. of
hydrogen + 1 vol. of oxygen.

EXPT. 38.— Electrolytic Gas.— If two

electrodes are placed in a bottle of dilute
acid (Fig. 41), the hydrogen and oxygen
gases come off mixed together in the form
of electrolytic gas. This is washed from
acid spray by a little water in the bulb
tube, and collected over water in a stout
soda-water bottle. This is wrapped in a
towel, and the gas ignited with a taper

(p. 51). A very violent detonation occurs (hence electrolytic gas is

sometimes called detonating gas).

EXPT. 39. — A thin glass flask is filled with the mixture and inverted
over a cork carrying two stout copper wires connected with a Ruhmkorff
coil (Fig. 42). The flask is covered with a cage of stout fine-mesh iron

FIG. 41.— Preparation of
Electrolytic Gas.

INORGANIC CHEMISTRY

CHAP.

FIG. 42. — Explosion of Electrolytic
Gas by an Electric Spark.

wire gauze and a spark passed. There is a violent detonation, and the
flask is shattered, a little finely -powdered glass escaping through the
gauze in the form of white smoke.

The Volumetric Composition of Water.

EXPT. 40. — Detonating gas is passed
into a stout graduated glass tube, with
sparking -wires above, filled with mercury,
and inverted in a trough of that metal
(Fig. 43). This tube is called a eudio-
meter. When a little gas has collected,
the eudiometer is held down firmly on a
pad of rubber moistened with mercuric
chloride solution, beneath the mercury.
On passing an electric spark, there is a
flash of light in the tube, accompanied
by a dull noise (not an explosion), and on raising the eudiometer,
mercury rushes in and fills it, with the exception of a few drops
of water which are
seen floating on the .
surface of the metal.
A more conveni-
ent form of eudio-
meter is shown in
Fig. 44. It consists
of a strong glass
U-tube filled with
mercury, graduated,
and provided with a
stopcock and firing
wires on one side ;
the other limb is a
plain open tube, with
a stopcock below for
running off mercury.
About 4 c.c. of elec-
trolytic gas are intro-
duced through the
stopcock, mercury
being run off from
the lower stopcock.

A large volume of

FIG. 43. — Eudiometer for Explosion of Gases by an Electric Spark.

out, so as to lower the pressure of the gas, the thumb is pressed firmly
over the open end of the tube, and the gas fired by a spark.

THE COMPOSITION OF WATER

EXPT. 41. — By using a U-shaped eudiometer, the graduated limb of
which is surrounded by a glass jacket through which the vapour of
boiling amyl alcohol (131-132°) is passed, the water produced by the
explosion is kept in the form of vapour (Fig. 45). Thirty c.c. of electro-
lytic gas are introduced, measured at the temperature of the jacket,
with the mercury levels adjusted to equality on both
sides by lowering the mercury reservoir. The open end
of the U-tube is firmly closed by the thumb, and a
spark passed from the coil. There is a flash of light,
and an immediate contraction when the thumb is
removed. By running mercury into the open limb until
the levels are again equal, it will be seen that the
residual steam occupies 20 c.c. The 30 c.c. of

FIG. 44.

U-shaped

Eudiometer.

FIG. 45. — Volumetric Composition of Stsam.

electrolytic gas contained, as we know, 20 c.c. of hydrogen and 10 c.c.
of oxygen, hence :

2 vols. of hydrogen -f- 1 vol. of oxygen = 2 vols. of steam.

The accuracy attained in these experiments is not sufficient to
give the exact figure for the combining volumes of the gases.
Cavendish's result, giving the ratio of the volumes of hydrogen
and oxygen uniting to form water as 201 : 100, is very near the
ratio 2:1, which was accepted until 1888. The more exact experi-
ments made since that date have shown that the ratio is probably

INORGANIC CHEMISTRY

CHAP.

very nearly 200-3 : 100, but a description of these experiments
is deferred until Chapter XII.

The composition of water by weight. — Since it is difficult to weigh
with accuracy large volumes of hydrogen and oxygen, it is only
comparatively recently that the composition of water by direct
synthesis from its elements has been attempted. Formerly an
indirect method was used. A stream of hydrogen, which is not
weighed, is passed over weighed copper oxide (prepared by heating
copper turnings in air) heated to dull redness. The oxide is reduced
by the hydrogen to metallic copper, the oxygen of the oxide uniting
with the hydrogen to form water, which is collected and weighed.
From these results we find :

Loss of weight of copper oxide = weight of oxygen = o.
Weight of water - weight of oxygen = weight of hydrogen = h.
.*. Ratio of combining weights = ofh.

Fia. 46. — Gravimetric Composition of Water.

It will be seen that the weight of hydrogen is obtained by difference,
so that the synthesis is not complete.

This method was applied by Berzelius and Dulong in 1819, who
obtained the ratio : oxygen : hydrogen : : 8'01 : 1, approximately.

EXPT. 42. — About 20 gm. of black oxide of copper, previously heated
to redness in a crucible and cooled in a desiccator over calcium chloride
to remove moisture, are introduced into a hard glass bulb -tube, A
(Fig. 46), which is then weighed. The tube is connected by a rubber
or ground glass joint to a small receiver, B, attached by a rubber stopper
to a U-tube, C, filled with granular calcium chloride, a substance which
readily absorbs water vapour (p. 203). The receiver, B, and tube, (7,
are weighed together.

A current of hydrogen, generated from pure zinc and dilute sulphuric
acid in the flask, D, and dried by the calcium chloride tube, E, is then
passed through the apparatus until all the air is expelled. The
gas bubbles out through sulphuric acid in the glass, F. The copper
oxide is then heated by a Bunsen flame. Drops of moisture at once

iv THE COMPOSITION OF WATER 61

condense in the lower part of A and in the receiver, B, and the black
oxide of copper is reduced to red metallic copper. B is kept cool in
a vessel of cold water, and as the experiment proceeds and A becomes
warm, all the water is driven over into B and (7. The apparatus is
allowed to cool, with hydrogen still passing. The tube A, and B and
C, are again weighed.

The gain in weight of B and C gives the weight of water formed. The
loss in weight of A gives the weight of oxygen given up by the copper
oxide to the hydrogen to produce this water. The difference between
the weight of the water and the loss of weight of the copper oxide gives
the weight of hydrogen.

In 1842 Dumas carried out this experiment with all the accuracy
possible at' the time. Hydrogen was generated from zinc and dilute
sulphuric acid, and was purified by passing through a train of seven
U -tubes containing : (1) lead nitrate solution to remove sulphuretted
hydrogen, (2) silver nitrate solution to remove arseniuretted
hydrogen, (3) three tubes of caustic potash to remove acid vapours,
(4) two tubes of sulphuric acid cooled in ice, or phosphorus pentoxide,
to dry the gas. The reagents were distributed on pumice or broken
glass to expose a large surface.

The copper oxide was contained in a large hard glass bulb with a
long neck. This was weighed after evacuation to remove the air.
The air was displaced from the apparatus by hydrogen, and the
bulb heated by a large spirit lamp for ten to twelve hours. The
water produced was collected in a smaller bulb, in the neck of which
calcium chloride was placed, followed by a series of four drying
tubes containing sulphuric acid on pumice, or phosphorus pentoxide.
The last tube communicated with a vessel of sulphuric acid, through
which the residual hydrogen escaped. In all the experiments
the weight of the last absorption tube was constant. The whole
apparatus is shown in Fig. 47.

The copper was allowed to cool in the bulb in a stream of hydrogen,
the hydrogen was displaced by air in the whole apparatus, and the
bulb then exhausted and weighed. The absorption system was
also weighed.

A mean of nineteen experiments gave the following result :

Percentage by Combining ratio

weight. by weight.

Oxygen .. .. 88-864 7-98

Hydrogen.. .. 11-136 1-00

100-000 8-98

This ratio was accepted without question for nearly half a

INORGANIC CHEMISTRY

CHAP.

century. Dumas himself, however, had pointed out two sources
of error in the method :

( 1 ) Air dissolved in the sulphuric acid passed on with the hydrogen,

-^C 3^

^^ v M 30

iltffsElffl

!5|_§2S

'.S^IIS

OS 0^_-

B«3a!W"B

^ 5«g«fl«i

"«« ^-cJ:g5gg|r:

31ill!ll8?

« 4^ cs 5 i" ^ *
75 G-g — ^ ^
p

iilIsiPi:

ft • "' 59 «•> •— <D "3

THE COMPOSITION OF WATER

and the oxygen of this air combined with hydrogen in the copper
oxide bulb ;

(2) The reduced copper retained hydrogen when cooled in
that gas. Both errors tended to reduce the loss of weight of
the bulb, so that the proportion of oxygen found would be too
small.

In 1890 Dittmar and Henderson found an additional error in
Dumas' method. In drying hydrogen with t
sulphuric acid, sulphur dioxide is formed.
When this is passed with hydrogen over
heated copper, the oxygen of the sulphur
dioxide combines with the hydrogen to
form water, whilst the sulphur remains in
combination with the copper as sulphide.
The loss in weight of the copper oxide bulb
is therefore seriously too small. By using
hydrogen dried with caustic potash and
phosphorus pentoxide, however, these ex-
perimenters found the still lower ratio
oxygen : hydrogen : : 7-93 : 1.

Reiser in 1888 introduced the method of
weighing the hydrogen absorbed in palladium
(p. 71) ; he weighed the water formed on
pumping the gas over heated copper oxide,
which was not weighed. Oxygen was found
by difference and the ratio was found to be
7-935-7-975 to 1. Noyes in 1890 burnt
hydrogen in a copper oxide bulb and con-
denser made in one piece, the increase of
which gave the weight of hydrogen. The
water was removed and its weight found.
The loss of weight of the apparatus gave
the weight of oxygen. Thus a complete
synthesis was effected, and the result was the
ratio 7-947 : 1.

The most exact experiments on the composi- FIG. 48.— MORLEY'S COM-

« -V , , ,, /. f~r TT7 BUSTION TUBE.

tion of water by weight are those of E. W. The gases passed through
Morlev (1895). Purified oxygen and hydrogen phosphorus pentoxide

' . T -T • , J& i i 1 &- drying tubes, b, b, to the

gases were weighed in large glass globes ; in jets, a, a, where they were
the later experiments the hydrogen was weighed SSn/^andT"0 SParkS
in a bulb of palladium. The gases were then
burnt at platinum jets in a previously evacuated sealed glass vessel
(Fig. 48), immersed in cold water, the gases being ignited by an
electric spark between the wires shown. The water was then frozen,
and the residual gas pumped out through a tube containing
phosphorus pentoxide (to keep back water vapour), and analysed.

64 INORGANIC CHEMISTRY CHAP.

A typical experiment furnished the following data :

Weight of hydrogen introduced into apparatus = 3-8223gm.

,, residual hydrogen = 0-0012 ,,

,, hydrogen burnt = 3-8211 ,,

„ oxygen introduced into apparatus = 30-3775 ,,

,, residual oxygen = 0-0346 ,,

oxygen burnt = 30-3429 „

Sum of weights of hydrogen and oxygen burnt = 34-1640 ,,

Weight of water produced = 34-1559 ,,

.'. Loss in weight due to experimental error = 0-0081 ,,
Ratio of weights of oxygen and hydrogen combining to form water
= 7-941 : 1.

As a final result, the mean of twelve experiments in which
400 gm. of water were produced, Morley obtained the ratios :

Oxygen : hydrogen :: 7-9396 : 1
Water : hydrogen :: 8-9392 : 1.

By his other series of experiments on the densities and combining
volumes of the two gases (pp. 72, 213), Morley found the ratio :

Oxygen : hydrogen :: 7-9395 : 1.

These researches are probably the most exact chemical investi-
gations ever executed.

SUMMARY OF CHAPTER IV

The formation of water on the explosion of a mixture of hydrogen and
air, or oxygen, was noticed by Priestley (1781). More exact experiments
of Cavendish (1781-1784) established the fact that almost exactly 2
vols. of hydrogen and 1 vol. of oxygen combine to form water, but the
clear statement that water is composed of these substances is due to
Lavoisier (1785). Nicholson and Carlisle (1801), and Cruickshank
(1802), found that water is decomposed into its elements by an electric
current (electrolysis), the hydrogen appearing at the negative pole and
the oxygen at the positive. No other substances are produced from
pure water (Davy, 1806).

The volumetric composition of water has been determined by exploding
measured volumes of the gases, and measuring the residual gas. Scott
(1887-93) found oxygen/hydrogen = 1 : 2-00285 ; ' Burt and Edgar,
in a very careful research (1915), found the ratio 1 : 2-00288 (p. 213).

The gravimetric composition of water was determined by : (1) passing
hydrogen over heated copper oxide ; (2) burning weighed quantities
of hydrogen and oxygen, and weighing the water. Dumas (1842), by
method (1), found : hydrogen/oxygen — I : 7-98 ; Dittmar and Hender-
son (1890) found errors in Dumas' method, and obtained 1 : 7-93.
Cooke and Richards (1887) found 1 : 7-934, Reiser (1888) 1 : 7-93. By
method (2) Rayleigh (1889) found 1 : 7-945, and Morley (1895) 1 : 7-9396.

iv THE COMPOSITION OF WATER 65

EXERCISES ON CHAPTER IV

1. Give a short account of the work leading to the discovery of the
composition of water, stating the share of each investigator in the eluci-
dation of the problem.

2 How would you proceed to illustrate by experiment the composi-
tion of water by weight and by volume ?

3. Describe any investigation in which the composition of water has
been accurately determined.

4. 21*40 gm. of lead oxide are heated in a current of hydrogen, and,
after reduction, the weight was 19-46 gm. What weight of water has
been formed ?

5. Describe experiments designed to produce (a) detonating gas,
(6) oxygen and hydrogen gases separately, from water. How may
the relative volumes of detonating gas and the steam produced from
it by explosion be compared, and what is the result ?

CHAPTER V

THE PHYSICAL PROPERTIES OF GASES AND VAPOURS

Compression of gases : Boyle's law. — The discussion of gaseous
pressure, and that of the effects of changes of volume and tem-
perature on the pressure of a gas, belong to physics. A brief
summary of the results, presented in such a form as to be imme-
diately applicable to chemical problems, may, however, be given
here.

Effect of pressure on volume. — Boyle's . law (1662) : When the
temperature is maintained constant, the volume of a given mass of gas
is inversely proportional to the pressure :

pro = constant — C (1)

The density of a gas is the mass per unit volume, m/v, hence
the density is proportional to the pressure. If we call the mass of gas
hi grams which occupies 1 c.c. its concentration, then at constant
temperature the pressure is proportional to the concentration.

Boyle's law is not exact ; all gases show marked deviations from
it at high pressures. At moderate pressures all common gases
except hydrogen are more compressible than an ideal gas which
obeys Boyle's law. Hydrogen is slightly less compressible, and
the same behaviour is shown by all gases at very high pressures
(Amagat).

Table of Relative Volumes occupied by various gases when 1 vol. at the

given pressure is reduced to atmospheric pressure. Temperature 16°.

50 100 120 150 200

atm. atm. atm. atm. atm.

Ideal gas 50 100 120 150 200

Hydrogen 48-5 93-6 111-3 136-3 176-4

Nitrogen 50'5 100-6 120'0 147-6 190-8

Air 50-9 101-8 121-9 150-3 194'8

Oxygen 105-2 -212-6

Do. atO° 52-3 107-9 128-6 161-9 218-8

Carbon dioxide 69-0 477* 485* 498* 515*

* Liquefied at pressures greater than 90 atm.
66

CH. v THE PHYSICAL PROPERTIES OF GASES AND VAPOURS 67

At very low pressures (0-01-1-5 mm. Hg) no deviation from
Boyle's law can be detected (Rayleigh, 1901-2). Boyle's law appears
to be exact under such conditions, and the gases behave like the
ideal gas.

Effect of temperature : Charles's law. — Dalton in 1801 observed
that gases expanded by equal increments of their volumes for
equal rises of temperature ; his results were published in 1802.
In the same year Gay-Lussac published a memoir, in which he
stated that Charles, in 1787, had found that gases expand equally
between 0° and 80°, but did not measure the expansion. Gay-
Lussac, from his own experiments, derived the law in question,
which differs from Dalton's in the reduction of the initial volume
to 0°. It is known as Charles's law : at constant pressure all gases
expand by 1/273 of their volume at 0° C. for a rise of temperature of 1°.

Let VQ — volume at 0°, vt = volume at t°, under the same pressure,

then vt = v0 (l -f J^\ or vt/v9 = (273 + 0/273. If vlt v.2 are the

V £ id/

volumes corresponding to two temperatures ^°, t2°,
vjvi = (273 + y/(273 + y.

The value (^ + 273) is called the absolute temperature, Tlt
corresponding to ^ ; hence, the volumes are proportional to the
absolute temperatures (p const.) : v2/v^ == T^T^. If we put
t= - 273, then T = 0, and by substitution in the equation we find
that v = 0. The temperature — 273° is called the zero of absolute
temperature, or the absolute zero. It can be proved by thermo-
dynamics that it is impossible to cool a body below the absolute
zero. By the rapid evaporation of liquid helium in a vacuum,
Kamerlingh Onnes obtained a temperature of —271-5°, or only
1-5° above the absolute zero.

If the volume of a given mass of gas is kept constant, the pressure
increase for 1° is 1/273 of the pressure at 0°. This is readily proved
from Boyle's and Charles's laws. Thus p2/Pi — T^T^.

If volume and temperature change together, it is readily shown
in the same way (cf. Duncan and Starling's " Text -book of Physics "
(Macmillan), p. 406) that : ^W/^i = ^2^2/^2 ; or> generally, pvjT =
constant, for a given mass of gas.

Charles's law is not strictly true ; the coefficients of expansion of
gases differ slightly among themselves, and from 1/273, and the change
of pressure at constant volume is slightly different from the change of
volume at constant pressure, for the same rise of temperature. At
very low pressures, however, these magnitudes approach equality, the
limiting value being 1/273-09. The exact value of the absolute tempera-
ture of melting ice is therefore 273-09°. For the ideal gas the coefficient
of expansion is 1/273-09 = 0-0036618.

F 2

68 INORGANIC CHEMISTRY CHAP.

EXAMPLE 1. Boyle's law. — A volume, of gas occupies 224 c.c. when
under a pressure of 755 mm. What will be its volume under a pressure
of 760 mm., if the temperature remains constant ?

The volume is inversely proportional to the pressure :

224 : i'2 : : 760 : 755,

= 224 X = 222-5 c.c.

Alternative method : — pv = constant .;, p^
,. ^ = ,^ = 224 x™ =222-5 c.c.

EXAMPLE 2. Charles's law. — 450 c.c. of gas are collected at a tempera-
ture of 16°. What will be the volume at 0° if the pressure remains
unchanged ?

The volume is proportional to the absolute temperature :

*!° C. = 273° + ^° abs. = Tf abs. = 273° + 16° = 2S9° abs.
t2° C. = 273° + t° abs. = T2° abs. = 273° abs.

v, : v2 : : T, : T2 /. v2 = v, X |? = 450 X |™ = 425'1 c.c.

EXAMPLE 3. Combined gas law. — A quantity of hydrogen at 15° and
750 mm. pressure occupies 4-5 litres : what will be its volume at 0°
and 760 mm. ?

wx£X-g- 4-5 x™x™ = 4-209 Htres.

The density of a gas. — The density of a gas, or vapour, is expressed
in two ways : —

(1) The normal density, or simply density, of a gas or vapour
is the weight in grams of 1 litre (or 1000-027 c.c.) of the substance,
measured at a temperature of 0°, and under a pressure of 760 mm.
of mercury, the weights being reduced to sea-level, and latitude 45°.

One litre is defined as the volume occupied by 1 kilogram of water,
at 4°, weighed in vacuum at sea-level, and latitude 45°. One cubic
centimetre is the capacity of a centimetre cube, the centimetre being
one-hundredth of the length of the standard metre. ^Owing to a slight
inconsistency in the Metric System, the volume of 1 gram of water at
4° is not 1 c.c., but 1-000027 c.c. Since weight is slightly variable with
the position on the earth, it is referred to standard conditions, sea-
level and lat. 45°.

(2) The relative density of a gas, or vapour, is the weight of any
volume of the substance divided by the weight of an equal volume
of pure hydrogen, measured and weighed under the same con-
ditions.

v THE PHYSICAL PROPERTIES OF GASES AND VAPOURS 69

Hydrogen is chosen as the standard substance because it is the
lightest gas known.

Standard temperature and pressure (or normal temperature and
pressure), denoted by S.T.P. (orN.T.P.), are 0° C. (273-09° absolute),
and the pressure of a column of 760 mm. of mercury at 0° at
sea-level, and at latitude 45°. On account of slight deviations of
gases from the laws of Boyle and Charles, the relative density in
accurate work is determined with both gases actually at S.T.P., so
that no correct:ons by the gas laws are necessary. With vapours
this is, of course, impossible, but, on account of the very large
deviations of vapours from the gas laws, an approximate value of
the relative density is all that is in this case determined and
required.

Determination of gas densities. — The density of a gas is deter-
mined by weighing an evacuated globe, filling it with the gas,
and reweighing. The volume of the
globe is determined by filling it with
water and reweighing.

EXPT. 43. — Fit a 1 litre round -bottomed
flask with a rubber stopper and glass
stopcock (Fig. 20). Evacuate the flask
by a good Fleuss or Geryk oil -pump
(Fig. 49), or a metal water-pump with a
calcium chloride tube attached to prevent
diffusion of moisture into the flask.
Weigh by suspending on one arm of a
large sensitive balance. Connect the
flask with a gas-holder of carbon dioxide,
interposing a calcium chloride tube.
Open the stopcocks on the gas-holder,
and slightly open the stopcock on the

evacuated flask so as to allow the gas to stream slowly into it.
When the pressure is equalised, run out water from the gas-holder
by the lower tap until the level of water in the funnel tube is the
same as that inside the gas-holder, and close the stopcock on the
flask. Reweigh the latter. Now fill the flask with water to the level
of the cork and weigh on a rough balance to find the volume of the flask
(assume 1 gm. = 1 c.c.). Read the barometer and the temperature of
the water in the gas-holder. Calculate the density of carbon dioxide at
S.T.P.

If the globe is weighed first vacuous, then full of the gas, and
finally filled with hydrogen under the same conditions, the

weight of gas filling globe

FIG. 49.— Air-pump.

relative density is given by.:

weight of hydrogen filling globe.

70 INORGANIC CHEMISTRY CHAP.

If the weighings are carried out under different conditions, the

density of gas (at S.T.P.)
relative density is the ratio : density of hydrogen (at S.T.P.).

EXAMPLE 1. — Weight of evacuated flask = 148-563gm.

„ ., flask filled with carbon dioxicle = 150-382,,
„ „ „ „ water = 1128-6 „

Temperature of gas = 15° ; pressure (barometer) = 758 mm.
Volume of flask = 1128-6 - 148-6 = 980 c.c. This volume is occu-
pied by the carbon dioxide at 15° and 758 mm. pressure,
.'. volume of carbon dioxide at S.T.P.

7KQ 97Q

= 980 x^X || = 926-6 c.c.

Weight of carbon dioxide = 150-382 - 148-563 = 1-819 gm.,
.*. density = weight of 1 litre (1000 c.c. approximately)
1-819 X 1000 - Q,jQ gm.
-926*- * litT

EXAMPLE 2. — Weight of above flask filled with hydrogen at 14° and
759 mm. pressure == 148-646 gm.

759 273
Volume of hydrogen at S.T.P. = 980 X ^ X ^ = 930-9 c.c.

Weight of hydrogen = 148-646 — 148-563 = 0-083 gm.

0-083 X 1000 n AftQ gm.
of hydrogen = - - 9

densitv of carbon dioxide
Relative density of carbon dioxide = densrty of hydrogen

1-963

0-089

= 22-05.

The true weight (in vacuum) of the globe is the apparent weight
in air plus the weight of air displaced by the globe : this latter
value depends on the temperature, pressure, and degree of moisture
of the air, and as these may be different during the separate weighings,
corrections of all weights to vacuum will be necessary in accurate
work. Also, the surface of the globe always carries a film of
moisture condensed upon it from the atmosphere (cf. p. 23),
which will vary with the moistness of the air. To eliminate these
difficulties as far as possible Regnault introduced the use of com-
pensating globes. The density globe was counterpoised on the
balance by hanging on the other arm another globe of as nearly as
possible identical weight and volume (Fig. 50), so that all variations
of atmospheric conditions affected both globes equally, and the
corrections were thus eliminated. The small adjustments of weight
necessary, corresponding with the weights of the gases themselves,

THE PHYSICAL PROPERTIES OF GASES AND VAPOURS

were made with ordinary metal weights, which are corrected to
vacuum in calibration, and in any case have a negligible displace-
ment.

A correction which remains to be made when this method is used
was pointed out by Rayleigh (1888), viz., that due to the shrinkage
of the globe on evacuation. This results in the globe displacing a
little less air when it is evacuated than when it is full of gas, or than
the compensating globe. The amount of shrinkage is found by
pumping out the globe in a closed vessel filled with water, and
observing the fall of level of the latter in a communicating graduated
tube. With a globe
of 2000 c.c. capacity
the correction to
be applied was
0-0006 gm. on the
weight of hydrogen
filling the globe, and
Regnault's value of
0-08968 for the nor-
mal density of hy-
drogen had to be
raised to 0-08988.

This method has
been used by E. W.
Morley (1896) in a
very careful deter-

FIG. 50.— Determination of the Density of a Gas. ,

mination of the
normal densities of
hydrogen and oxy-
gen, which are of
fundamental im-
portance in chem-
istry (p. 122). Very
pure hydrogen gas
was absorbed in a
glass tube contain
ing metallic palla-
dium, which is capable of taking up considerable amounts
of hydrogen, but not of gaseous impurities, so that the latter
may be removed by pumping out the tube. On heating the
palladium to dull redness, pure hydrogen is evolved from the
metal, and the loss in weight of the tube gives the weight of gas.
The hydrogen was received in three large evacuated glass globes,
immersed in ice, the total volume of the globes being accurately
known. The rise in pressure in the globes was then determined by a
mercury manometer. One result is given below.

72 INORGANIC CHEMISTRY CHAP.

Volume of the three globes 43-2574 litres

,, „ „ gas space in manometer 0-0550 litre

,, „ „ connecting tubes 0-0365 ,,

Total volume of gas 43-3489 litres

Temperature 0°. Pressure 725-40 mm. Loss of weight of palladium
bulb = weight of hydrogen = 3-7164 gm.

Correction to reduce weighings to sea- level and latitude 45°, and
length of cathetometer to 0° = 1-00044,

normal density of hydrogen

3-7164 ' 760
43*489 X 725*

0-089861 gm./litre.

litre

As a mean of all his results, Morley found :
Normal density of hydrogen == 0-089873 ± 0-0000027 gm. per
Normal density of oxygen = 1-42900 ± 0-000034 gm. per litre.

In comparing the first figure with the corrected result of Regnault,
Morley's weighings must be reduced to the latitude of Paris. His
value then becomes 0-089901, differing from Regnault's, 0-08988,
by less than 1 in 4000.

The f ollowing table gives the most recent values of normal densities
of gases, determined with great exactness :

Air ..

1 -2928'

Nitrous oxide

1-9777

Hvdrojjen

chloride

1-6398

Oxygen ...

... 1-42906

Nitric oxide

1-3402

Sulphur

dioxide

2-9266

Hydrogen

... 0-08987

Ammonia

0-7708

Helium...

0-1782

Nitrogen

... 1-2507

Carbon

monoxide . . .

1-2504

Neon ...

0-9002

Argon

... 1-7809

Carbon

dioxide

1-9768

Methane

0-7168

The relative density of air is 1-2928 -r 0-08987 = 14-44. Formerly,
densities of gases were referred to air = 1 instead of to hydrogen = 1 ;
these values may be converted to the modern units by multipli-
cation by 14-44. The composition, and therefore the density, of
air vary slightly in different localities, hence the use of this gas as
an accurate standard of relative density is strongly to be deprecated
(p. 536).

The law of partial pressures. — If two or more gases, which do
not react chemically, are mixed together in a closed vessel, the
pressure exerted by the mixture of gases is the sum of the pressures
which each gas alone would exert if separately confined in the whole

v THE PHYSICAL PROPERTIES OF GASES AND VAPOURS 73

volume occupied by the mixture. (The temperature is assumed to
be maintained constant throughout.) The pressures exerted by the
separate gases are called their partial pressures, and the above
statement is called the law of partial pressures. (Dalton, 1801.)

EXPT. 44. — Connect two globes, A and B (Fig. 51), of capacities
about 2 and \ litres, respectively, with each other and a manometer
as shown. Close the stopcocks T2 and T3 and partially evacuate A
through the cock T±. Close T4 and establish connection with the
manometer by opening Ty Read the difference in mercury levels,
and subtract from the reading of the barometer to find the
pressure of the gas. Let the
pressure in A be pA mm. In
the same way, reduce the pres-
sure in B to ptt mm. Close
T3 and "open Tl and Tz. When
the two quantities of air have
mixed, and the temperature has
regained the initial value, open
T3 and read the final pres-
sure, p. Total volume = VA -f- vs,
.*. partial pressures of the air
in A and B, respectively, are

and

-, these,

V* + VB VA -f- VB

by Boyle's law, being the pres-
sures the separate quantities of
air would exert if each occupied
the whole volume VA + VB. The
sum of the partial pressures is

'• — B, and this will be found to be very nearly equal to p.

FIG. 51. — Experiment on the Law of
Partial Pressures.

In one experiment the following results were found : —
Volume of large flask = Vj, = 2210 c.c.
Volume of small flask — VB = 600 c.c. ,

Pressure of gas in large flask .= 76 — 20 cm. = 56 cm. mercury
Pressure of gas in small flask = 76 cm. of mercury = pa.

2210

2810

Partial pressure of first gas in mixture = 56 X ^~~ = 44 cm.

Partial pressure of second gas in mixture =

Observed total pressure after mixing = 76 — 16 = 60 cm.
Sum of partial pressures = 44 + 16-2 = 60'2 cm.

(Air was used in both flasks.)

74 INORGANIC CHEMISTRY CHAP.

The law of partial pressures is not strictly exact ; all real gases
show slight deviations from it. Leduc has shown that the law given.
below is more exactly followed than the law of partial pressures :
the volume occupied by a mixture of gases is equal to the sum of the
volumes which the component gases would occupy at the same temperature,
and under the same pressure, as the mixture. This has been verified
with mixtures of hydrogen and nitrogen up to 200 atm. pressure.
EXAMPLE 1. — Two vessels, of capacities 500 c.c. and 2000 c.c., contain-
ing hydrogen and oxygen, respectively, under pressures of 750 mm.
and 10 mm., are put in communication. What will be the final
pressure of the mixture of gases ?

Total volume = VA + VB = 500 c.c. + 2000 c.c.
/. partial pressure of hvdrogen

VA ' 500

= P* x^T+^ = 750 X2500 = 150mm"
and partial pressure of oxygen

VB 2000

-P. X ^-pTB = 10 X 2500= 8 mm'
Total pressure = sum of partial pressures = 150 -f- 8 = 158 mm.

Alternative method : sum of partial pressures = A^ _^_ ^B '
(750 X 500) + (10 X 2000)

~2000 + 500~

EXAMPLE 2. — 154 c.c. of nitrogen, at 750 mm. pressure, are mixed with
50 c.c. of hydrogen, at 550 mm. pressure, in a vacuous globe of capacity
2000 c.c. What is the partial pressure of hydrogen in the mixture, and

the total pressure ? •**

-7w/>^LA*. 154

Partial pressure of hydrogen = 750 x T = 57'75 mm., inde-

pendently of the presence of the other gas in the vessel.

Partial pressure of nitrogen in mixture = 550 X 2000 = ^'75 mm.

/. Total pressure = sum of partial pressures = 57-75 +.15-75 =
73*5 mm.

Vapour pressure. — Liquids when admitted to vacuous spaces
evaporate, or give off vapour, until the latter attains a definite
pressure, which depends only on the temperature. ^ The vapoury
is then said to be saturated, Dalton's law of partial pressures shows/
that the pressure of the vapour of a liquid in a closed vessel filled
with an indifferent gas will also be the same as if the space were
initially vacuous. If insufficient liquid be present to saturate
the space, the vapour is said to be unsaturated.

EXPT. 45. — Fill two tubes about 78 cm. long, sealed at one end, and
carefully cleaned and dried, with dry mercury, and invert in two small
dishes containing mercun'. The level of the mercury sinks in each

v THE PHYSICAL PROPERTIES OF GASES AND VAPOURS 75

tube, leaving a vacuous space above. Measure the level of mercury in
each tube above the surface in the trough.

By means of small pipettes (Fig. 52) introduce a few drops of water
into one tube, and a few drops of ether into the other. Notice the
depression of the mercury in both cases, and that the
effect due to the ether is much greater than that caused
by the water. Measure the levels again, and find the
vapour pressures of the two liquids at the atmospheric
temperature. Warm the tube containing ether with the
hand and notice the further fall of the mercury, due
to the increase of vapour pressure with temperature.

The vapour pressure of a liquid rises very rapidly
with the temperature. This is evident from Fig. 53, r
which is the vapour pressure curve of water. When
the vapour pressure becomes equal to the total [
pressure exerted on the surface of the liquid, say
by the atmosphere,, the liquid boils, i.e., vapour is
emitted in bubbles throughout the whole bulk of
the liquid. The boiling point of a liquid is the tem-
perature at which its vapour pressure becomes equal vapou?Pressure
to the atmospheric pressure, or other total pressure, vacuous1 acea
acting on the surface of the liquid. Boiling points are
usually given for a pressure of 760 mm., or 1 standard atmosphere.
If the pressure on the surface is reduced, say by connecting a
flask containing the liquid with a vacuum pump, the boiling

point is depressed. Thus, under a

row\»ua, pressure of 17-4 mm., water boils

at 20°. It is therefore necessary
to specify the pressure in giving a
boiling point ; unless this is done,
it is understood that the pressure
is 760 mm. Thus, the above
result would be expressed as :
20°/174 mm. The boiling point
of a pure liquid may be used as
a means of characterising the

200

substance (cf. p. 3).
EXPT. 46.— The effect of

^ 20 30 «, *, 60 70 so so w
FIG. 53.-V.pour Pressure Curve of Water. °n ^boiling point may be shown

by boiling water in a strong round-
bottomed flask, corking the flask, and placing it in cold water. Owing
to the condensation of steam in the upper portion of the flask the
pressure is reduced, and the water boils vigorously. This experiment
is due to Bishop R. Watson.

INORGANIC CHEMISTRY

CHAP.

The vapour pressure of a liquid is the same in a vacuum as in
a space filled with an indifferent gas.

EXPT. 47. — Place a small sealed thin glass bulb, containing 2 c.c.
of bromine (Fig. 54) inside a 500 c.c. bottle. Fit a rubber stopper
to the bottle, through which pass a glass tube, closed at one end,
and with the other end over the point of the bulb below, and a
small manometer, containing mercury. Depress the tube so as to
fracture the bulb, and observe the rise of pressure indicated by
the manometer. Notice the formation of a layer of red bromine
vapour in the lower part of the bottle. This diffuses upwards and
the pressure rises as the space becomes saturated.

Vapour pressures of solids. — Not only liquids but also solids
possess definite vapour pressures at different temperatures. These
are usually smaller than those of liquids, although solids may have,

at a given temperature, greater vapour
pressures than liquids of different com-
position.

EXPT. 48. — Pass a small piece of cam-
phor into the vacuous space in the
barometer tube (Expt. 45) surrounded
above by a hot-water jacket, arid notice
the fall of the mercury. Determine the
vapour pressure of camphor, and compare
it with that of water at the same
temperature.

The vaporisation of solids without
previous fusion is called sublimation.

Equilibrium. — At a given temper-
, ature, liquid (or solid) and vapour

FIG. 54. — Vapour Pressure of a Liquid ••_/*• i A -.L i •

in a space filled with Gas. can exist indefinitely in contact when
the pressure of the vapour is equal

to the maximum vapour pressure at that temperature. The
vapour is then saturated. Under these conditions the system
composed of the two phases, liquid and vapour, is said to be
in equilibrium. An equilibrium state is"1 one which is independent
of time. If we represent transition from liquid to vapour by the
symbol : [Liquid] -> [Vapour], i.e., evaporation, and transition from
vapour to liquid by : [Vapour] -> [Liquid], i.e., condensation, the
state of equilibrium will be represented by [Liquid] ^± [Vapour],
or, more concisely, Liquid ^± Vapour.

Moist gases. — In the laboratory, gases are often collected over
water, and if an accurate measurement of the volume of the gas is
to be made, it is necessary to correct for the water vapour it contains.

v THE PHYSICAL PROPERTIES OF GASES AND VAPOURS 77

If water evaporates into a dry gas at constant pressure, the gas will
expand. The volume of a given mass of gas is therefore greater
when it is moist than when it is dry.

Suppose we have a volume of 100 c.c. of moist air, measured over
water at 15°, and under a total pressure of 760 mm. This total pressure
is, by the law of partial pressures, the sum of the pressure of the dry air
and of the maximum vapour pressure of water at 15°, viz., 12-7 mm.
The pressure of the dry air is therefore 760 — 12-7 = 747-3 mm. If
the water vapour were removed by a drying agent from the 100 c.c.
of moist air contained in a closed vessel, the pressure would therefore
fall to 747-3 mm. If we now increased the pressure of the dry air to

747-3

760 mm., the volume would become, by Boyle's law, 100 x c.c.,

and at 0° this would be 100 X

747-3

273
X - = 93-2 c.c.

In general, if a mass of gas saturated with moisture at t° under
.a total pressure of P mm. occupies F c.c., the volume of dry gas
at S.T.P. will be :

P-f 273

760 K 273 + t °'C''
where / is the vapour pressure of water at t°.

If partially saturated gases are measured over mercury, they may be
saturated with water vapour by introducing a few drops of water into
the measuring tube. This only applies, of course, to gases which are
not appreciably soluble in water.

In using this formula we require a table of the vapour pressures
of water at different temperatures. A portion of such a table is
given below.

TABLE OF

VAPOUR

PRESSURES OF WATER.

Temp.

Vapour
pressure
in mm.
mercury.

Temp.

Vapour
pressure
in mm.
mercury.

Temp.

Vapour
pressure
in mm.
mercury.

Temp.

Vapour
pressure
in mm.
mercury.

0°

4-569

17°

14-39

30°

31-51

93°

588-3

6-534

15-33

54-9

610-6

9-140

16-32

92-0

633-7

9-77

17-36

148-9

657-4

10-43

18-47

233-3

681-9

11-14

19-63

354-9

707-1

11-88

20-86

525-5

733-2

12-67

22-15

545-8

100

760-0

13-51

23-52

566-7

110

1075

INORGANIC CHEMISTRY

CHAP.

Intermediate values in the practically useful ranges 0-25° and
90-100° may be obtained by interpolation, such as is used with logarithm
tables.

EXAMPLE 1. — Find the vapour pressure of water at 15-4°.

Vapour pressure at 15° = 12.- 67 mm. Vapour pressure at 16° =
13-51 mm.

/. difference for 1° = 13-51 — 12-67 = 0-84 mm., .*. difference for
0-4° = 0-84 X 0-4 = 0-34 mm. /. vapour pressure at 15-4° = 12-67
4- 0-34 = 13-01 mm.

EXAMPLE 2. — Find the volume, dry and at S.T.P., of 175 c.c. of air
measured over water at 18° and 749 mm. atmospheric pressure.
V = 175 ; P = 749 mm. ; / = 15-33 mm. (from table) ; t = 18°.

749 - 15-3 273

.'. required volume = 175 X -

760

x 27~3

s ss sa

Humidity. — The weight of aqueous
vapour contained in a given volume
of moist air, divided by the weight
which would be contained in the same
volume of saturated air, is called the
hygrometric state, or the humidity, of the
moist air. Methods of determining
humidity are described in the text-
books on physics, and we shall merely
refer here to what is known as the
chemical method.

In this a known volume of air
is drawn by means of an aspirator

through a weighed series of U -tubes containing calcium chloride,
or pumice soaked in sulphuric acid, which absorbs the moisture
(Fig. 55). The weight of moisture in a given volume of air is
thus found. Now it is known that the weight of 1 litre of
aqueous vapour at S.T.P. (if it could exist at that temperature
and pressure, and followed the gas laws) is 0-7962 gm. From the
hygrometric experiment we should have found, however, that

FIG. 55. — Determination of Humidity
by the Chemical Method.

v THE PHYSICAL PROPERTIES OF GASES AND VAPOURS 79

x grams of aqueous vapour were contained in 1 litre of the air,
and if we divide this by the amount contained when the air is
saturated at t°, found from the table below, we obtain the humidity.
If the partial pressure of aqueous vapour in the air under given
conditions is/', the weight in grams of water in 1 litre will be, at t°,

A.7QA9 v I f V

< V760 X 273 + t,

since the expression in brackets is the volume of vapour in litres
at S.T.P., and the expression outside is the weight in gm. of 1 litre
of aqueous vapour under these conditions.

If f = f, the maximum vapour pressure, we obtain the table
given below. It is easily seen that the humidity is given by the
ratio /'//.

Weight of Water Vapour in Grams in i Litre
of Saturated Air

Temp.
°C

Weight
of
Vapour

Temp.
°C

Weight

of
Vapour

Temp.

°c

Weight
of
Vapour

Temp.

°c

Weight
of
Vapour

Temp.
°C

Weight
of
Vapour

•0049
•OO52

•0182

•0533

•1348

•3073

•0056

•0193

•0560

•1407

•3128

•0060

•0204

•0588

•1468

•3246

•0064

•0216

•0618

•1532

•3368

•0068

•0228

•0648

•1597

•3493

•0072

•0241

•0681

•1666

•3623

•0077

•0255

•0714

•1736

•3756

•0082

•0270

•0749

•1809

•3894

•0088

•0285

•0785

•1885

•4035

•0094

•0307

•0823

•1963

•4180

•0100

•0317

•0862

•2044

•4330

•0106

•0335

•0902

•2127

•4454

•0013

•0353

•0945

•2213

•4643

•0120

•0372

•0989

•2302

•4806

•0127

•0393

•1034

•2395

95.

•4974

•0135

•0413

•1082

•249O

•5146

•0144

•0435

•1131

•2588

•5323

•0152

•0458

•1183

•2689

•5505

•0162

•0482

•1235

•2794

•5693

•0171

•0507

•1291

•2901

100

•5884

The figures in the table have been calculated by the equation above.

80 INORGANIC CHEMISTRY CHAP.

The average humidity of the air in London during January is
0-7. Hence, when the temperature is 0°, the weight of moisture
in 1 litre of such air is 0-0049 x 0-7 = 0-00343 gm.

The law of partial pressures applied to vapours is not exact ; the
vapour pressure of a liquid in a gas is slightly less than in vacuo. It is
only at low pressures, i.e., at low temperatures, when the vapour pres-
sures are small, that the application of the "gas laws to vapours, made in
the preceding equations, is justified. This is very nearly the case at
the ordinary atmospheric temperature.

Density of a moist gas. — It may also be necessary to find the
density of moist air (or other gas) of a given saturation at a given
temperature. Consider 1 litre of moist air at t°, under a pressure
P mm., and let/' be the partial pressure of aqueous vapour in the
air. f = saturation pressure at t° (from table of vapour pressures)
X humidity. The volume of the dry air at S.T.P. will be :

and since the weight of 1 litre of dry air at S.T.P. = 1-2928 gm.
(p. 72), the weight of the dry air will be

1-2928 X
The volume of aqueous vapour at S.T.P. will be :

and since the (hypothetical) weight of 1 litre of aqueous vapour at
£>.T.P. is 0-7962 gm., the weight of the aqueous vapour in the 1 litre
of moist air will be :

°-7962 x 4 x 2TO gm-

The total weight of the litre of moist air will therefore be :
1-2928 X

1 1 -2928 (P -/') +0-7962 /' [> gm.

(273 + t) 760

which is the density under the given conditions.

The same calculation applies to other moist gases, the appropriate
density being used in place of 1 -2928, the value for air.

EXAMPLE. — Find the weight of 1 litre of hydrogen, saturated with
moisture at 15°, and under a pressure of 740 mm.

v THE PHYSICAL PROPERTIES OF GASES AND VAPOURS 81

Normal density of hydrogen = 0-08987 gm. per litre (p. 72) ; vapour
pressure of water at 15° = 12-67 mm., hence required weight

°'08987 (74°-]3) + °'7962 x 12'67

= 0-09417 gm.

Note that, whereas moist air is lighter than dry air, the reverse is the
case with hydrogen. This is because aqueous vapour is lighter than
air but heavier than hydrogen.

Vapour densities. — Since vapours when far removed from their
points of liquefaction obey approximately the same laws of expan-
sion as gases, it is possible, if the weight of a known volume of vapour
is determined at a given temperature and pressure, to reduce this
volume to S.T.P. and so find the normal density of the vapour.
This will be a hypothetical value, since the substance cannot
really exist under such conditions, but it is the most convenient
value for comparative purposes. The ratio of this number to the
weight of 1 litre of hydrogen at S.T.P. is the vapour density, usually
denoted by A.

The vapour density may also be found by dividing the weight of
any volume of the vapour measured under the actual temperature
and pressure of the experiment by the weight of an equal volume
of hydrogen measured and weighed under the same conditions.

The weight of V c.c. of hydrogen at a temperature t° and under
a pressure of P mm. is :

The vapour density of a volatile liquid or solid may be deter-
mined by one or other of the following methods ; that selected
in any particular determination depends on the conditions of
experiment, e.g., whether a high or low temperature, or pressure,
is used : —

(1) Hofmann's modification of Gay-Lussac's method : volume of a
given weight of vapour is found.

(2) Dumas' method : weight of a given volume of vapour is
determined.

(3) Victor Meyer's method : volume of air displaced by a given
weight of vapour is determined.

Hofmann's method.— A. W. Hofmann (1868) surrounded a
barometer tube with a glass jacket through which the vapour of a
liquid boiling in a separate vessel was passed. Uniformity of
temperature was thus assured. The liquid is weighed into a small
bulb (Fig. 56) with a ground stopper, which is forced out under

INORGANIC CHEMISTRY

CHAP.

the diminished pressure when the bulb is passed into the upper
part of the barometer tube. The latter is a wide tube, at least
1 metre in length, carefully graduated (Fig. 57). The
liquid rapidly vaporises in the vacuous space above
the mercury in the barometer tube ; the bulb, of
course, must be completely . filled with the liquid,
since a bubble of air will expand considerably in the
vacuous* space. The vapour jacket is provided with a
side tube near the bottom for leading the vapour to the
condenser.

FIG. 56.
Bulb for
Liquid.

The following liquids may be used in the boiler for supplying the
vapour -jacket, the particular liquid taken depending on the boiling

FIG. 57. — Hofmann's Method for Determination of Vapour Density of a Liquid.

v THE PHYSICAL PROPERTIES OF GASES AND VAPOURS 83

point of ths substance examined : the boiling points under 760 mm.

pressure are stated :

Water, 100° Toluidine, 202°

Amyl alcohol, 131-132° Ethyl benzoate, 212°

Ariiline, 181-5° Amyl benzoate, 261°

Since volatilisation occurs more readily under diminished pressure,
steam may often be used for the jacket in determining the vapour
densities of liquids which boil under full atmospheric pressure as high
as 180°. If the atmospheric pressure during the experiment differs
appreciably from 760 mm., the boiling point of the liquid furnishing
vapour to the jacket must be corrected by using special tables, or a
thermometer may be hung in the vapour jacket.

When the mercury level is constant, the following data are noted :

(i) The volume of the vapour in c.c. = V.

(ii) The temperature, t°, in the jacket.

(iii) The pressure of the vapour ; this is approximately given by
the barometric height, H mm., minus the height of mercury in the
tube above the level in fhe trough, h mm. ; i.e., (H — h) mm.

In accurate work, the height of the heated mercury in the column in
the tube must be reduced to 0°, to correspond with the corrected baro-
meter reading, and allowance made for the expansion of the scale
of the glass tube. The vapour pressure of mercury at the tem-
perature of the jacket is also subtracted from the pressure of the
vapour.

Let the weight of substance taken be m grams. The weight of a
volume of hydrogen equal to that of the substance under the con-
ditions of the experiment is

FX 0-00009
The vapour density, A, is then m\m ' .

EXAMPLE. — 0-338 gm. of carbon tetrachloride gave 109-8 c.c. of vapour
in a Hofmann apparatus, at 99-5°. Barometric height = 746-9 mm.
Height of mercury in tube above level in bath = 283-4 mm.

070 7d.fi. Q _ 9&3«d.

. m' = 109-8 x 0-00009 X — g X - —t -^ - = 0-0044 gm,
. vapour density of carbon tetrachloride = 0-338/0-0044 = 76'8.

Dumas' method.— The method invented by Dumas (1827) is
an extension of that commonly used for permanent gases (p. 69).
Since the vapour does not come in contact with mercury, the
method may be applied to substances (e.g., bromine) which cannot

G 2

INORGANIC CHEMISTRY

FIG. 58.

Dumas' Vapour
Density Bulb.

be dealt with by Hofmann's method, and it may also, by the use of
porcelain globes, be used at higher temperatures. It is not so
accurate as the former method, and as the vaporisation is
carried out under atmospheric pressure, and the temperature of
the vapour is higher, it cannot be used for substances which
readily decompose.

In Dumas' method a glass bulb (Fig. 58) of about 200 c.c.
capacity, with a drawn-out neck, is cleaned, dried,
and weighed. By warming the bulb, dipping the neck
in the liquid to be examined, and cooling, sufficient
liquid is introduced into the bulb to expel all the
air when it is volatilised.

The bulb is then supported in an iron pot containing
water, oil, or melted paraffin wax, heated 30-40°
above the boiling point of the liquid, so that only
the ^P °* tne ^ulb Projects above the surface of
the liquid in the bath (Fig. 59). Volatilisation
rapidly occurs, the air being expelled from the
globe, and the vapour is at
a temperature sufficiently above
the boiling point to obey the
gas laws with fair approxima-
tion. When the rush of vapour
ceases, the neck of the globe is
sealed off, and the temperature
of the bath read off on the
thermometer.

The globe is removed from the
bath, cooled, cleaned, and re-
weighed along with the piece
of neck sealed off. The neck is
then scratched with a file, and
the tip broken off under the sur-
face of previously boiled water.
The latter rushes into the bulb
and, if the experiment has been
successful, fills it completely.
The bulb full of water is
weighed, together with the two
small pieces of the neck. The
barometric pressures during the

second weighing, and at the time of sealing, are noted.
Let the weight of the globe in air = m gm. ;

weight of the globe filled with vapour = m1 gm. ;
weight of the globe filled with water = ra2 gm.
The volume of the globe = ra2 — m c.c.

FIG. 59. — Dumas' Vapour Density
Apparatus.

v THE PHYSICAL PROPERTIES OF GASES AND VAPOURS 85

The weight of air filling the globe at the temperature t, and
pressure h, when it is weighed full of vapour, will be :

070 -L

(m, -m) X 0-001293 X g^p X ^ gm. = A gm,

hence the weight of the vacuous globe in air = m — A gm., and
weight of vapour filling the globe = m' — (m — A) gm.

The weight of an equal volume of hydrogen at the temperature
t' and pressure H of sealing will be :

(m, -m)x 0-00009 x -, X = «'

'/. Vapour density A = (ml~ (m — A)} / m.

In some cases the weight of vapour may be found by chemical
methods. E.g., if iodine has been used, the tip of the bulb is broken
off under potassium iodide solution, which dissolves the iodine, and the
solution is then titrated with sodium thiosulphate (p. 522).

EXAMPLE. — The vapour density of hexane.

Weight of empty globe in air = 23-449 gm. ;

„ „ globe and vapour at 15'5° = 23-720 gm.

Temperature of sealing 110°; barometric pressure 759 mm.,
unchanged throughout the experiment. Capacity of globe, by weighing
water, 178 c.c.

Weight of air displaced by globe

070 7p;Q
= 178,, X 2|1^- X ^ x 0-001293 - 0-2175 gm.,

/. weight of vacuous globe = 23-449 — 0-218 = 23-231 gm.,

„ vapour = 23-720 — 23-231 = 0-489 gm.
Weight of hydrogen filling globe at 110° and 759 mm.

070 7KQ

= 178 x x~ x o-00009- = 0<0114 m-

Vapour density A = 0-489/0-0114 = 43'8.

The chief drawbacks to Dumas' method are the large quantity
of substance required to displace the air of the bulb and the circum-
stance that, if the substance contains impurities of higher boiling
point, these come off last and render the vapour sealed up impure,
the density being too high.

Deville and Troost (1860) extended Dumas' method to higher
temperatures by using globes of porcelain, heated in the vapours
of mercury (357°), sulphur (444-6°), stannous chloride (660°),
cadmium (778°), or zinc (918°), in an iron bath (Fig. 60), and sealing
off the* tip of the bulb with the oxy-hydrogen blowpipe. To find
the temperature of the globe a companion globe filled with iodine,
the density of which had been determined at various temperatures,

INORGANIC CHEMISTRY

CHAP.

was placed alongside the other globe. In this way the variation
of the vapour density of many substances, e.g., sulphur, with
temperature (cf. p. 150) was found.

Victor Meyer's method. — Several new methods of finding vapour
densities were devised by Victor Meyer, the most useful being the
so-called displacement method (1878). This method is more rapidly
and easily carried out than those of Dumas and Hofmann, requires
only a small quantity of the substance, and gives quite accurate
results.

A long glass tube with a bulb, b (Fig. 61), and a side tube, a, is
heated in a long vapour bath, c, at a temperature which must be
constant and higher than the boiling point of the substance, but
need not otherwise be known. The tube a delivers into a graduated
tube, gr, in a trough of water. The tube b is heated in the bath
until no more bubbles of air escape from a ; then the latter is placed

under the graduated tube, the cork,
d, at the top of the long tube is taken
out, and a weighed quantity of the
liquid hi a small stoppered bulb
dropped into the heated bulb, the
cork being quickly replaced. A little
asbestos is placed in the bottom of
the bulb, 6, to prevent fracture on
dropping in the bulb of liquid. It
is also more convenient to drop in
the bulb through a large bore stop-
cock instead of the cork at d.

The substance quickly vaporises,
and the vapour, which does not diffuse
to the top of the narrow tube, dis-
places its own volume of air, which is collected in the graduated
tube. When no more bubbles come off, the water levels in the
tube and trough are equalised and the volume of air is read off.

Let the volume of moist air at the temperature t° of the trough,
and under a barometric pressure H, be V c.c. If the vapour
pressure of water at t° is/ mm. (cf. table on p. 77), the volume of
dry air at S.T.P. wiU be :

97O ff f

FX27i^X^C-C-=F°C-C-

This is the volume which the vapour of the given weight of sub-
stance would occupy at S.T.P. if it could exist under these con-
ditions. The weight of an equal volume of hydrogen is 0 -00009 F0
gm., so that if m gm. of substance were used, we have simply :

Vapour density A = ra/0 -00009 F0.

FIG. 60. — Vapour Densities at High
Temperatures (Deville and Troost).

v THE PHYSICAL PROPERTIES OF GASES AND VAPOURS

EXAMPLE. - - 0-1008
gm. of chloroform ex-
pelled 20-0 c.o. of moist
air at 15° and 770 mm.
pressure. Vapour pres-
sure at 15° = 13 mm.

/. volume of dry air
atS.T.P.

273 770 — 13

= 20 X

288 760

= 18-9 c.c.

Weight of an equal
volume of hydrogen
= 18-9 X 0-00009 gm.
= 0-00169 gm.

.*. vapour density _of
chloroform =
0-1008/0 -00 169 = 59'6.

Victor Meyer's method
is not suitable for sub-
stances which break up
on heating, and decom-
pose still further when
under reduced pressure
(e.g., phosphorus penta-
chloride, p. 153), since,
owing to admixture of
the vapour with air in
the bulb, the partial
pressure of the vapour
is reduced to an extent
which is not known.

The following sub-
stances may be used in
the heating bath : water
(100°), amyl alcohol
(132°), xylene (140°),
aniline (181- 5°), ethyl
benzoate (212°), benzo-
phenone (306°), di-
phenylamine (310°),
mercury (357°), sulphur
(444-6:), molten lead.

FIG. 61. — Victor Meyer's Vapour Density Apparatus.

INORGANIC CHEMISTRY

CHAP.

Apparatus for

at High Tem-

Measurements by Victor Meyer's method at high temperatures
were made by Nilson and Pettersson (1889), and later by Biltz
and V. Meyer, who used bulbs of glazed porcelain, protected by

wrapping them with thick
platinum foil, placed inside
graphite crucibles heated in
a Perrot's gas furnace up
to 1730°. By using water-
gas in the furnace the tem-
perature reached 1900°. The
bulb is filled with inert gas (nitrogen,
or argon) to prevent chemical action,
and the substance, weighed out in a
glass bulb, is dropped in as usual.
Nernst (1903) used a small iridium
bulb (3 c.c.), painted outside with
zirconia, and heated electrically to
2000° in a small iridium tube. The
substance (usually a fraction of a milli-
gram) was weighed on a micro-balance
sensitive to 1/2000 mgm., and the dis-
placement measured directly by the
movement of a drop of mercury in the
horizontal graduated side tube (Fig. 62).
A more sensitive micro-balance, sensitive to 1/500,000 mgm.,
was used by Ramsay and Gray (1911) in determining the density
of radium emanation, 0-1 cu. mm., or less than 0-001 mgm., being
used. It consists (Fig. 63) of
a beam, A, of quartz rods,
10 cm. long, weighing 0 -3 gm. ,
with a quartz knife-edge, B,
resting on a polished quartz
plane, C. A small pan, D,
and a sealed bulb, E, of
known volume, both of
quartz, are suspended from
one end of the beam by a
quartz fibre, and are coun-
terpoised by a bead of
fused quartz, G, on the
other end of the beam.
Oscillations are observed

by a beam of light reflected from a mirror, H, through a
glass window, K, in the air-tight metal case, M, on a scale
several yards away. Weighings are made by altering the
pressure of the air inside the balance case, and so changing

FIG. 62. — Nernst's A
Vapour Densities
peratures.

FIG. 63. — Micro- balance.

v THE PHYSICAL PROPERTIES OF GASES AND VAPOURS 89

the buoyancy of the bulb, E. The pressure is measured by a
manometer, P.

SUMMARY OF CHAPTER V

Physical properties of gases : — 1. Boyle's law (1662) : when the. tem-
perature is constant the volume of a given mass of gas varies inversely as the
pressure : vl : v2 : : p2 : pv or pv = const.

2. Charles's law : all gases expand by the same fraction, ^g-, of their
volume at 0° C. for 1° rise in temperature, when the pressure is constant :

Vt == v0 (l + ^A If we put T = t° + 273, the absolute tempera-
ture, the volumes at constant pressure (or the pressures at constant volume)
are proportional to the absolute temperatures : v^ :v2: : T±: T2. By com-
bining this with Boyle's law we find : pvIT = const.

3. The normal density of a gas is the weight in grams of 1 litre (1000-027
c.c.) at 0° and 760 mm. pressure (standard temperature and pressure =
S.T.P.), the weight being reduced to sea-level and latitude 45°. The
relative density of a gas (or vapour) is the ratio of the weight of any
volume of the gas to the weight of an equal volume of hydrogen, under
the same conditions. One c.c. of hydrogen at S.T.P. weighs 0-00009 gm.

4. Dalton's law of partial pressures : if two or more gases, which do not
interact chemically, are mixed in a vessel, the pressure of the mixture is
the sum of the partial pressures , i.e., the pressures which would be exerted
by each component if separately confined in the whole space occupied by the
mixture.

This applies (approximately) to vapours : the vapour pressure of a
liquid is the same in a vacuum as in a space filled with an indifferent
gas, and depends only on the temperature.

5. Vapour densities are determined by : (1) Hofmann's method
(volume of a given weight found) : (2) Dumas' method (weight of a
given volume found) ; (3) Victor Meyer's method (volume of air dis-
placed by the vapour from a given weight of substance found under
atmospheric conditions).

EXERCISES ON CHAPTER V

1. A volume of gas occupies 50"c.c. when measured over water at
15°. The barometric pressure is 747 mm. Find the volume of the
dry gas at S.T.P. If the gas is oxygen, what would be its weight ?

2. A hydrogen cylinder of 2 cu. ft. capacity is filled by compression
to 200 atm. If the gas is used in filling a balloon at atmospheric
pressure, what volume will pass into the balloon ?

3. Two hundred c.c. of hydrogen and 50 c c« of nitrogen, each
measured at 15° and 760 mm.,' are admitted in succession to a pre-
viously exhausted 500 c.c. flask. What is the pressure of the mixture at
18°? '

4. Describe the method used in an accurate determination of the
density of a gasv Explain the terms normal density and relative
density as applied 'to gases. How may the relative density referred to
hydrogen = 1 be converted into that referred to air = 1 ?

90 INORGANIC CHEMISTRY CH. v

5. What methods are in use for the determination of vapour densities,
and what are the advantages and disadvantages of each V Describe
one method in detail.

6. In the determination of the vapour density of a substance by
Dumas' method, the following data were obtained :

Weight of bulb in air = 44-7832 gm. Weight of bulb and vapour
filling it at 115° = 45-1848 gm. Weight of bulb filled with water =
234-0 gm. Temperature of balance case = 12-8°. Barometric height =
75-1 cm. Find the vapour density.

7. In Victor Meyer's method it was found that 0-323gm. of alcohol
expelled 171-2 c.c. of air measured over water at 15-2° and 76-29 cm.
Find the vapour density of alcohol.

8. What is meant by the humidity of air ? It was found that 10
litres of air at 14-8° and 750 mm., when aspirated through calcium
chloride tubes, caused an increase of weight of 0-1036 gm. Calculate :
(i) the weight of 1 cu. m. of the moist air ; (ii) the humidity.

9. Find the weight of 1 litre of hydrogen, saturated with water vapour
at 15°, under 740 mm. pressure. If the pressure of the hydrogen is
doubled, what is the weight of 1 litre of the moist gas ?

CHAPTER VI

SOLUTIONS AND THE PHASE RULE

Equilibria between the phases of water. — It has been explained
(p. 74) that if a quantity of liquid water is contained in a closed
space it gives off vapour until a definite pressure is attained,
for each temperature, known as the vapour pressure. When the
vapour has attained the vapour pressure corresponding with the
particular temperature, the liquid and vapour will exist together
indefinitely, and are then said to be in equilibrium. This state of
equilibrium between the two phases (p. 7) of water is denoted by
the symbol : Water (liquid) ^ Water (vapour).

In the same way, ice and water may co-exist in equilibrium at
a particular temperature and pressure ; thus, at 0° and under a
pressure of one atmosphere (1-033 kgm. per sq. cm.) the two phases
remain in contact without change for any length of time : Water
(liquid) ^± Water (solid). Since the vapour pressure of water varies
with the temperature, we should expect the temperature at which
ice and water co-exist also to be influenced by pressure, or, in other
words, that the melting point of ice will depend on the pressure.
It is found tiiat +hf* TYiAlfjng pr>in+. rvP i^o ic. fofrffrfry fry pressure.

EXPT. 49. — Hang a wire, carrying heavy weights at its ends, over a block
of ice supported on a trestle. The wire gradually cuts its way through
the block, since the ice melts beneath the wire, but the ice remains intact
after the wire has passed through, because the water freezes again when
the pressure is released on the ice becoming liquid.

The melting point of ice is lowered by 0-0073° by each additional
atmosphere pressure. If snow is pressed between the hands it
will cohere to a snowball unless it is very cold. In the latter case
the degree of pressure which can be exerted by the hands cannot
lower the melting point to the external temperature. Thus, if the
snow is at — 10°, a pressure of 10/0-0073 = 1370 atm. would be
required. But in a hydraulic press a mass of transparent ice is
formed. The fusion of ice under pressure, and its resolidifi cation
when the pressure is taken off, is called regelation.

92 INORGANIC CHEMISTRY CHAP.

Water in contact with vapour and water in contact with ice are
heterogeneous systems, each composed of two phases. We denote
the number of phases in a system by P.

Degrees of freedom. — We shall now explain what is understood
by the number of degrees of freedom of a system, denoted by F.

If we have 1 gm. of water vapour at 100° confined in a cylinder
under a piston, and if the pressure on the piston is always kept
below 1 atm. or T033 kgm. per sq. cm., the vapour will behave more
or less like a gas. (At very low pressures it will behave like an
ideal gas : pressure x volume = const.) At all pressures less
than 1 atm. the vapour is homogeneous, and at a given pressure and
temperature it has a definite volume which, since 1 gm. has been
taken, is called the specific volume, v. Now if we have the water
vapour at a fixed temperature, under a fixed pressure, and with a
fixed specific volume, it is completely defined. E.g., it will have a
definite heat capacity, a definite thermal conductivity, refractive
index, etc. But if any two of the three independent variables,
pressure, temperature, and specific volume, are fixed, the state will be
completely defined, since the third variable will assume automatically
a definite value. The same holds good if we have 1 gm. of liquid
water in the cylinder. We express these results by saying that a
homogeneous system composed of a pure gas or liquid has two degrees
of freedom ( F = 2), since two of the three variables may be arbitrarily
fixed before the state is completely defined.

If the temperature of the water vapour remains at 100°, but the
pressure is increased above 1 -033 kgm. per sq. cm., then liquid water
appears. The pressure now becomes constant, and remains equal
to 1-033 kgm., as the piston descends, because the only effect of
reducing the total volume is to cause vapour to turn into liquid.
During this process the two phases (liquid and vapour) have con-
s^tant temperatures and specific volumes, and each is under a
constant pressure ; hence they are both completely defined, arid the
system is in equilibrium. The only variable left is the temperature ;
if this is changed, the pressure and the specific volumes alter. In
the same way, if to a mixture of ice and water we apply pressure,
ice will melt, but the pressure and the specific volumes remain
constant. A heterogeneous system of two phases (P = 2) of a pure
substance has one degree of freedom (F = 1). The same is true if
we have solid ice in contact with water vapour below 0°.

The triple point — Ice, liquid water, and water vapour can exist
together in equilibrium (with fixed specific volumes) only at one
particular temperature (0-0077°), and under one particular pressure
(4-57 mm. mercury). The heterogeneous system of three phases
(P = 3) possesses no degree of freedom (F = 0), since it is com-
pletely defined only when all the variables, pressure, temperature,
and specific volumes, are fixed. ' This state is known as the triple

vi SOLUTIONS AND THE PHASE RULE 93

point, and is defined by the values of the temperature and pressure :
tf^O'00770, 2? = 4-57 mm. Other pure substances (acetic acid,
benzene) have different triple points.

Solutions. — We know that various kinds of natural water exist,
such as rain water, river water, and sea water, which show different
properties. If a natural water, e.g., ordinary tap water, is evapor-
ated to dryness in a platinum dish a white residue is left, showing
that the water contained solid matter in solution. The residue
from sea water is much larger than from the other forms of water,
and consists mainly of common salt. The peculiar properties of sea
water are due to the
dissolved salt. Thus
liquids, such as water,
can hold solids in solu-
tion.

Dissolved solids are
separated from liquids
by the process of
distillation, A simple
apparatus for distil-
lation consists of a
glass retort with the
neck passing into a
glass flask, or receiver,
which is cooled by a
stream of cold water
(Fig. 64). If tap water
is distilled in this ap-
paratus, the mineral
matter remains in the
retort, and distilled
water collects in the
receiver. If larger
quantities of liquid are
to be distilled it is
more convenient to use
a Liebig's condenser (Fig. 65), consisting of a glass tube enclosed in
a jacket through which a constant stream of cold water is passed.
The liquid to be distilled is contained in a distilling flask, provided
with a side tube which is passed through a cork in the condenser.
In the neck of the distilling flask a thermometer is supported
by a cork, so as to enable the boiling point of the liquid to be
determined.

It is possible by means of distillation to separate not only solutions
oi solids in liquids, but also, at least partially, solutions of liquids
in liquids. Thus, if a mixture of equal volumes of alcohol (b. pt.

FIG. 64. — Retort and Receiver arranged for Distillation.

94 INORGANIC CHEMISTRY CHAP.

78-3°) and water (b. pt. 100°) is distilled, it is found that the boiling
point at the commencement of the operation is 84°. The liquid
collecting in the receiver is richer in alcohol than the original
mixture, and will burn when lighted in a dish. As the distillation
proceeds, the boiling point rises, and the distillate contains more
and more water. If the distillation is stopped when one-fourth
of the mixture has distilled over, and the boiling point has risen to
85; 5°, it will be found, if the distillate is poured into a clean flask
and the operation repeated, that it begins to boil at 81 '5°, i.e., at a
lower temperature than the original mixture, and the first portion
of the distillate is correspondingly "richer in alcohol. This partial

FIG. 65. — Distilling Apparatus with Liebig's Condenser.

separation of a solution of liquids by interrupted distillation is
known as fractional distillation.

If a flask and delivery tube are completely filled with tap water,
and the flask is heated, bubbles of gas appear, which pass out of
the delivery tube under water (Fig. 66), and will be found to be
mainly air ; such water therefore contains dissolved gas.

Thus, liquids may hold in solution gases, liquids, and solids.

Solids are capable of dissolving gases ; thus palladium dissolves
hydrogen (p. 71), forming solid solutions. Solids may also dissolve
solids. Thus, if a piece of gold-leaf is pressed on a freshly-scraped
piece of lead, the gold slowly penetrates into the latter, as may be
proved by scraping off successive layers after a long time and
analysing them. Many gem stones (ruby, sapphire, emerald)
contain traces of metallic oxides, to which they owe their colour,
in solution in a transparent, colourless mass of other substances.

True solutions are homogeneous (p. 9), and the dissolved sub-

SOLUTIONS AND THE PHASE RULE

stance is in an extremely fine state of subdivision. Thus, 1 gm.
of eosin gives a distinct fluorescence (p. 8) to 1,000,000 c.c. of
water when examined in a strong light. Each cubic centimetre
of the solution contains only 0-000,001 gm. of the dye, and since
a volume of only 10~12 c.c. of solution can be examined under the
microscope, this can contain only 10" 18 gm., or

0-000,000,000,000,000,001 gm. of dye.

Colloidal solutions, such as that of arsenic trisulphide (p. 12),
pass through filter papers, and do not settle out on standing
as suspensions of larger particles do ; their heterogeneous character,
however, is disclosed by the ultra-microscope. Colloidal solutions
thus stand halfway between suspensions (separable by filtration)
and true solutions (homogeneous even under the ultra-microscope).
The radius of the particles of the dissolved substance in a true
solution must be of the order of
10-8 cm. (cf. p. 9).

The substance present in larger
amount in a solution, or the one
which has the same physical
state as the solution, is called the
solvent ; the other substance is
called the dissolved substance, or the
solute. Thus, a mixture of alcohol
and water may be called a
" solution of alcohol in water," or
a " solution of water in alcohol,"
according as water or alcohol is in
excess, but a very concentrated
solution of sugar in water, con-
taining more sugar than water,
is always called a " solution of sugar in water," because water
has the same physical state as the solution.

Solutions of gases in liquids : Henry's law (1803).— Solutions pi
gases (and vapours) in gases have already been considered in
Chapter V, pp. 72, 74. The law of partial pressures applies to
these cases, and asserts that the relation of the pressure to the con-
centration of one gas in a mixture is the same as if the other
gases were not present.

Solutions of gases in liquids may be studied by the apparatus
shown in Fig. 67, called an absorptiometer.

The gas is measured in the burette, A, over mercury, and the
volume reduced to S.T.P. Part of the gas is then passed into the
absorption vessel, B, the volume being found from that of the water
run out, and is shaken with the liquid until the solution
is saturated, i.e., until the equilibrium [Gas] ^ ['Dissolved Gas] is

FIG.- 66. — Expulsion of Dissolved Air
from Water.

INORGANIC CHEMISTRY

established. The absorption vessel is then placed in a bath of water
at a constant temperature, and the pressure adjusted by the level-
ling tube, C. The contraction in volume is then read off on the
burette, and since the volume of water which was left in B is known,
the number of c.c. of gas, reduced to S.T.P., which saturate 1 c.c.
of water (or other solvent, e.g., alcohol) can be calculated. This is
called the absorption coefficient. The water used in the experi-
ment must previously have been boiled to expel dissolved air, and
cooled in a corked flask.

If the gas is very soluble (e.g., ammonia, hydrochloric acid) it is
bubbled through a measured volume of water until the latter is
saturated. The amount of gas dissolved is then found by chemical
analysis (e.g., titration).

From the results of such experiments it has been found that the

amount of gas dissolved by a
fixed volume of liquid depends
upon (1) the chemical composition
of the gas and of the liquid,
(2) the temperature, (3) the pres-
sure. The effect of pressure is
given by Henry's law (1803) : the
amount oi gas absorbed by a liquid is
proportional to the pressure.

Since, however, the volume of a
given amount of gas is inversely
proportional to the pressure, it
follows that a given volume of liquid
absorbs the same volume of gas at all
pressures. One c.c. of water absorbs

0-0489 c.c. of oxygen at 0° and
FIG. 67.-Absorptiometer. ^ ^ . ^ ^^ ^.^ ^

volume at 2 atm., or 2 X 760 mm., if the gas is measured at S.T.P.
But this volume of gas still occupies 0-0489 c.c. under 2 atm. pressure.

Solubility of a mixture of gases in a liquid. — If we have to deal
with the solubility of a mixture of gases in a liquid, the amount of
any one gas dissolved is proportional to its partial pressure, when the
gas has come into equilibrium with the liquid. This is Dalton's
extension of Henry's law.

EXAMPLE. — The absorption coefficients of oxygen and nitrogen
in water at 0° are 0-04890 and 0-023481, respectively. One hundred
vols. of air contain 79-04 vols. of nitrogen and 20-96 vols. of oxygen.

20*96
Hence the partial pressure of the oxygen is -- = 0-2096 atm., that

SOLUTIONS AND THE PHASE RULE
79-04

of the nitrogen - = 0-7904 atm., the sum of these being the total

1 (.)( i

pressure, 1 atm. The volume of oxygen dissolved in 1 c.c. of water
when agitated with air at 0° under 1 atm. pressure will thus be
0-2096 X 0-04890 = 0-010244 c.c., and the volume of nitrogen dissolved
will be 0-7904 x 0-023481 = 0-018559 c.c. If the dissolved gas is now
expelled by heating, its composition will thus be 0-010244 vol. of
oxygen -f 0-023481 vol. of nitrogen, i.e., it will contain 64 '4 per cent,
of nitrogen and 35-6 per cent, of oxygen, by volume. It is therefore
richer in oxygen than the original air. By shaking this gas again with
water, and expelling the gas, the latter will be still richer in oxygen,
until after eight repetitions a gas containing 97-3 per cent, of oxygen is
obtained.

If the partial pressure of a gas above its solution be reduced to
zero, all the gas will be expelled from the solution. This can usually
be effected : (1) by reducing the pressure above the solution by an air-
pump ; (2) by passing a stream of indifferent gas through the solution
(e.g., nitrogen through aqueous ammonia) ; or (3) by boiling the
solution, when the dissolved gas is driven off with the steam.

In some cases it is impossible to remove all the gas by boiling,
etc. ; this occurs when the gas and solvent evaporate together to
form a vapour of the same composition as the solution ; the latter
then evaporates as a whole. Cf. p. 237.

Table of absorption coefficients. — Henry's law does not hold for
very soluble gases, such as ammonia at the ordinary temperature,
or hydrochloric acid, in water. It does not hold accurately for
carbon dioxide. At 100° the solubility of ammonia follows the law.
At higher pressures, also, deviations occur ; with more soluble
pises these begin at 2 atm. pressure, with less soluble gases the law
holds up to about 10 atm. A few absorption coefficients are given
below, in c.c. at S.T. P., absorbed by 1 c.c. of water under a pressure
of TOO mm.

0°

10°

15"

20°

30°

40°

50°

60°

Ammonia liSOO

910

so-j 1 710

—

Hydrochloric

acid ;JOG

474

458 442 411

386

362

339

Chlorine —

3-00

2-63

2-26

1-77

1-41

1*20

1-0

Carbon dioxide . . 1-713

1-194

1-019 0-87S 0-66

0-53

0-44

0-36

Oxygen O-04'.t

0-038

0-034 0-031

0-026

0-023

0-021

0-019

Nitrogen 0'023!'

0-0106

0-0179

0-0164 0-0138

0-0118

0-0106

o-oioo

Hydrogen <Mi-_'i;,

0-0198 0-0190 0-0184 ! —

98 INORGANIC CHEMISTRY CHAP.

Solutions of liquids in liquids. — Some liquids, such as water and
mercury, are practically, though probably not absolutely, immiscible ;
others, such as water and sulphuric acid, are completely miscible.
In some cases, such as ether and water, each liquid dissolves a limited
amount of the other, and the liquids are said to be partially miscible.
If successive small quantities of ether are added to water, they at
first dissolve completely. At a certain point, t^e water becomes
saturated with ether ; 100 gm. of water then take up 2-16 gm. of
ether at 22°. If more ether is added, a lighter layer separates,
and floats on the water solution. This is not pure ether, but
contains 11 gm. of water per 100 gm. of ether. With further
addition of ether (if the layers are shaken together), the composition
of each layer remains constant, but the lower (aqueous) layer
gradually disappears as more and more ether is added, until it
finally vanishes, the whole liquid then having the composition of
the upper layer. Unlimited further quantities of ether may now be
added without any separation of the homogeneous liquid into
layers.

The two liquid layers may be separated in a separating funnel
(Fig. 15) ; the presence of ether in the lower aqueous layer may be shown
by heating it in a test-tube, when the ether vapour given off may be
kindled. The presence of water in the upper ether layer may be shown
by dropping a bit of sodium into it, when hydrogen is evolved. (Pure
ether has no action on sodium.)

The compositions of liquid layers in equilibrium at 22° is given below.

Subst. in 100 Water in 100

gm. of water. gm. of subst.

Ether 2-16 gm. 11-02 gm.

Chloroform 0-64 „ 0-10 „

Carbon disulphide 1-24 „ 0-13 „

The partition law. — If to the two layers of ether and water a little
iodine is added, which dissolves in each pure solvent, it is found on
shaking that the iodine is shared between the two liquids, but
most of it, as is seen from the colours of the solutions, is taken by
the ether. In such cases, where a solute is shared between two
partially miscible or immiscible solvents, the ratio of the solute
concentrations in each layer is constant at a particular temperature,
independently of the absolute amounts of solute and liquids, or
the relative amounts of the two layers. (The concentration of a
solution is the weight of solute in 1 c.c. of solution.) The constant
ratio is called the ratio of distribution, or the partition coefficient,
and the result just stated, due to Berthelot and Jungfleisch (1872),
is called the distribution or partition law.

vi SOLUTIONS AND THE PHASE RULE 09

if Cj, c2 are the concentrations in the two layers, respectively, then:

=?= const. = k, or c, — /cc2,
where k is the partition ratio.

Thus, at 25° an aqueous solution of iodine containing O0516 gm. per
litre is in equilibrium with a solution of iodine in carbon tetrachloride
containing 4-412 gm. of iodine per litre.
The partition coefficient is :

concentration in carbon tetrachloride _ 4-412 _ „ r

concentration in wafer 0 • 051 0

A saturated solution of iodine in water at 25° contains 0-340 gm. per
litre. From the partition coefficient we can calculate the concentration
of a solution of iodine in carbon tetrachloride in equilibrium with a
saturated solution in water. Tin's is 0-340 X 85-5 = 29-1 gm. per litre.

Solutions of solids in liquids. — The most important class of solu-
tions is formed by dissolving solids in liquids. Common salt
added in successive small amounts to water dissolves up to a
certain point ; after this no more salt passes into solution, but settles
out unchanged. A solution which has dissolved as much solute
as is possible under the given conditions (e.g., at a fixed tempera-
ture) is called a saturated solution ; it can exist in equilibrium
with excess of solute : Salt [solid] ^ Salt [dissd.] . The concen-
tration of a solution of a solid in a liquid is expressed in
various ways ; usually as the number of grams of solute contained
in 100 gm. of solvent. The concentration of a solution saturated
with a solute is called the solubility of the latter ; it is the maximum
weight in grams of solid dissolved by 100 gm. of solvent at the
given temperature, in presence of the solid salt. For common salt
in water it is 35-9 at 15°. The solubility depends (1) on the chemical
character of the solute and solvent, (2) on the temperature, and
(3) to a slight extent on the pressure, in some cases (sodium chloride)
increasing, in other cases (ammonium chloride) decreasing, with
increase of pressure. The solubility of solids in water varies from
that of such " insoluble " substances as barium sulphate, to that
of very soluble substances such as calcium chloride. The very
small solubilities of such solids as barium sulphate have actually
been measured (p. 103).

The solubility usually increases with the temperature. In some
cases, such as sodium chloride, it is nearly independent of tem-
perature, and in others, such as calcium sulphate above 40°, it
diminishes with rise of temperature.

EXPT. 50. — The diminution of solubility with rise of temperature
may be shown by placing a tube containing calcium butyrate solution,
saturated at the ordinary temperature, in a beaker of boiling water.
In a short time crystals of the salt separate. These redissolve on cooling.

H 2

100

INORGANIC CHEMISTRY

CHAP.

The dependence of solubility on temperature is most conveniently
represented graphically by means of solubility curves, in which the

FIG. 68.— Solubility Curves.

abscissae represent temperatures, and the ordinates the solubilities.
In Fig. 68 the solubility curves of some salts in water are exhibited,
It will be seen that these show a great diversity

vi SOLUTIONS AND THE PHASE RULE 101

Supersaturation. — If a saturated solution of a salt is evaporated,
so as to drive off some solvent, and then allowed to cool, the salt
will be present in amount greater than corresponds with saturation.
Solid salt is then usually deposited in definite forms called crystals.
Many salts crystallising from aqueous solutions form crystals of
definite composition containing water. These are chemical com-
pounds, called hydrates. The water of hydra tion is usually known
as water of crystallisation. In many cases, however, when the
solution after boiling is cooled,, the salt is not deposited. The
cooled solution then contains more salt than corresponds with
saturation at the given temperature, and is said to be supersaturated.
Crystallisation is at once induced by throwing a small crystal of
the solid into the solution. It is found that large crystals are
deposited when the crystallisation takes place slowly. The solution
should not be too concentrated, and should be left at rest. Very
large crystals of alum, for example, may be obtained by allowing a
solution, saturated at the ordinary temperature, to evaporate
slowly in the air, and suspending a small perfect crystal in the
solution by a thread. If a very strong, hot solution is cooled
rapidly, and stirred, small crystals are deposited. These are usually
purer than the large crystals, since they are less likely to include
liquid. Mother liquor is removed by pressing with a spatula on
filter-paper in a Biichner funnel under reduced pressure. The mass
is then pressed with filter-paper on a porous plate of unglazed
earthenware, and finally dried in the air.

EXPT. 51. — Heat on a water-bath 250 gm. of crystallised sodium
thiosulphate (" hypo ") in a conical flask, the neck of which is plugged
with cotton-wool. The salt melts in its own water of crystallisation,
and forms a very concentrated solution. On cooling, this remains
liquid ; it is then a supersaturated solution. Remove the plug and drop
into the liquid a crystal of hypo. The liquid at once begins to solidify,
and the nmsa becomes warm, since heat is evolved in the process.
Supersaturated solutions invariably crystallise in contact with the
solid form of the solute.

EXPT. 52. — Fuse some crystals of hypo in a long test-tube, and pour
over the liquid a supersaturated solution of sodium acetate, prepared
by wanning the crystallised salt with one-quarter its weight of water
in a flask. Care should be taken to avoid mixing the liquids.
Plug the tube with cotton -wool and allow to cool. Remove the plug
and drop in a crystal of hypo. This falls through the acetate
solution without inducing crystallisation, but on its reaching the hypo
solution it brings about crystallisation of the latter. Now drop in
a crystal of sodium acetate. The upper liquid then crystallises.
Supersaturated solutions are caused to crystallise only by contact

102 INORGANIC CHEMISTRY CHAP.

with the particular solid dissolved in them (or a solid isomorphous. with
"

If a supersaturated solution is strongly cooled it may crystallise
spontaneously, without contact with solid. Thus, if a supersaturated
solution of hypo is cooled in a freezing mixture of Glauber's salt and
concentrated hydrochloric acid, it crystallises spontaneously.

Determination of solubility. — The solubility of a salt at various
temperatures is best determined by stirring the powdered solid
salt with water at the given temperature, so that excess of solid is
present, withdrawing a portion of the clear solution, weighing it,
and then evaporating the solution in a weighed dish to find the
weight of solid salt contained in it.

EXPT. 53. — A 25 c.c. pipette is fitted with pieces of rubber tubing
at each end. The upper piece of rubber is closed by a clip, whilst
the lower piece is connected with a short piece of tubing, packed with
glass wool, to serve as a filter (Fig. 69). The pipette is cleaned and
dried. A quantity of powdered potassium nitrate is now stirred with
water in a 250 c.c. beaker, set in a water-bath, and the temperature kept

FIG. 69.— Pipette for Determination of Solubility.

at 20°, as shown by a thermometer in the beaker. The mixture of
salt and water is best stirred by a small glass stirrer driven by a motor.
When the mixture has been stirred for fifteen minutes, the filter is
attached to the pipette, and solution drawn into the latter till the mark
is reached. The filter is detached, and the solution run from the pipette
into a weighed weighing bottle. The latter is stoppered, allowed to
cool, and weighed. The solution is poured into a weighed porcelain
dish, and any crystals are removed from the bottle, and washed into
the dish, by hot water. The liquid is then evaporated on a sand-bath
(spirting being avoided), and the dry residue weighed. The experiment
is repeated at 30°, 40°, and 50°, and the solubilities, in grams per 100 gm.
of water, plotted against the temperatures (Fig. 68).

Table of solubilities. — The solubility depends on the character of
the solid phase in contact with the solution ; in particular, salts
crystallising with different proportions of water have different
solubilities.

Thus, calcium sulphate occurs in two forms containing water.
One of these is the mineral gypsum. If gypsum is heated to 120-130 °
it loses part of its water of crystallisation, forming plaster of Paris.

vi SOLUTIONS AND THE PHASE Rl'LH 103

The latter is more soluble than gypsum, and forms with water a
solution supersaturated with respect to gypsum. The latter is
deposited, and more of the lower hydrate passes into solution.
This goes on until the whole has solidified to a mass of interlacing
crystals of gypsum (" setting " of plaster of Paris).

It will be seen from Fig. 68 that the solubility curves of sodium
and potassium chlorides are straight lines ; in general, however, the
solubility increases more rapidly than the temperature.

The great variation in solubility exhibited by different salts is
shown by the following table, giving the weights of anhydrous salt,
i.e., salt free from water, saturating 100 parts of water.

Salt. 0° 15° 100°

Potassium iodide 127-5 140 208

bromide 53-5 62-5 104

chloride 27-6 32-4 56-7

Sodium chloride 35-7 35-9 39-0

Calcium chloride (CaCl2,6H2O) ... 60 30°100 60°137

Strontium chloride (SrCl2,6H.2O) 43 50

Barium chloride (BaCl2,2H2O) ... 31-6 '34-4 58-8

Potassium nitrate 13-3 25-8 246

Sodium nitrate 73-0 85 178

Barium hydroxide (Ba( OH )2,8H2O) 1-67 3-23 101-4(80°)

Calcium hydroxide 0-185 0-170 0-077

Calcium sulphate (CaSO4,2H2O)... 0-18 40°0'279 60°0'260

Strontium sulphate , 0-0011

Barium sulphate 0-00023

Silver chloride 0-00015

„ bromide 0-00001

„. iodide 0-0000003

The freezing points of solutions. — If salt, sugar, or any soluble
substance is added to water, the freezing point of the latter is
depressed, and for each salt the depression is proportional to the con-
centration of the solution. (Blagden, 1788.)

Sea water freezes at about — 2° ; Bishop R. Watson (1771) experi-
mented with solutions of salt, exposing them to cold air : " in equal
quantities of water were dissolved quantities of sea salt, increasing in
the arithmetical progression, 0, 5, 10, 15, 20, etc. ; the times in which
the solutions began to freeze, reckoning from the time in which simple
water began, increased accurately in the same progression : hence it
may be inferred, that, in salt of the same kind, the resistance to congela-
tion is in the direct simple proportion of the quantity of salt dissolved :
this conclusion cannot be extended to salts of different kinds, since

104

INORGANIC CHEMISTRY

CHAP.

water saturated with sea salt is more difficultly congealed than when
saturated with various other salts, which it dissolves in greater
quantities."

The solid separating when solutions freeze is usually pure ice : all
the solute remains in the still liquid portion. Thus, freezing serves
to separate the constituents of a solution, just as distillation enables
us to achieve the same end (p. 93). The solution remaining will
become increasingly richer in salt as more and more ice separates,
and hence, by Blagden's law, the freezing point falls more and
more as ice continues to be formed.

When the freezing point has fallen to a certain value, it becomes
constant, and the whole of the solution then solidifies without

further fall in temperature, both ice and
solid salt then separating together in
the proportions in which they exist
in the solution. This minimum tem-
perature was formerly called tho
cryohydric temperature ; the mechanical
mixture of ice and solid salt separating
was supposed to be a compound, and
called a cryohydrate (Guthrie, 1874).
Later experimenters showed, by micro-
scopic examination and in other ways,
that the supposed cryohydrates are
mechanical mixtures, and they are now
known as eutectics, the lowest tempera-
ture attainable on freezing the solution
being called the eutectic point. In the
case of common salt and water this
is - 22°.

EXPT. 54. — The depression of freezing
point by a dissolved substance may be
shown in a lecture by the apparatus of
Fig. 70. An air-thermometer bulb is placed
in a large test-tube supported in a beaker.

The tube of the air thermometer has two bulbs blown on it, and dips
into coloured water. Water is first placed in the test-tube and a freezing
mixture in the beaker. On stirring, the water freezes, and the height
of liquid in the thermometer-tube is marked. One-hundred c.c. of an
aqueous solution of 6 gm. of urea are now used. The liquid rises
higher in the tube.

Vapour pressures of solutions. — If small drops of water are succes-
sively introduced into the vacuous space of a barometer- tube, sur-
rounded by a water-jacket to keep the temperature constant, the

FIG. 70. — Depression of Freezing
Point.

vr SOLUTIONS AND THE PHASE RULE 105

following effects will be noticed. At first the level of the mercury
falls after the addition of each drop, showing that vapour is being
produced in the space, and is exerting a pressure (p. 74). After a
certain number of drops have been added, liquid appears floating on
the surface of the mercury, and the pressure then remains perfectly
constant, even if more water is added. In the homogeneous vapour
phase we can have different pressures at a given temperature ;
there are two degrees of freedom. As soon as liquid appears together
with vapour, the pressure becomes constant, and equal to the
maximum vapour pressure. In the system composed of two
phases in equilibrium: Water [liquid] ^± Water [vapour] there is
only one degree of freedom. Now let a little common salt be passed
into the tube. It dissolves in the water, and the vapour pressure
of the latter is found to be slightly diminished. By adding further
successive quantities of salt, the vapour pressure falls still further.
Here we have a system composed of two phases, solution and
vapour, in equilibrium, which shows two degrees of freedom
instead of only one, as in the case of pure wfater. To what is the
difference due ?

It arises from the fact that the liquid phase is no longer a pure
substance which has a definite vapour pressure at a given tempera-
ture, but is a solution of variable composition, the vapour pressure
of which, just as the freezing point, depends on the concentration
of dissolved substance.

By adding more salt, the vapour pressure falls progressively
until a point is reached when the solution is just saturated with
salt. The vapour pressure is once more constant, since further
addition of salt has no effect on the concentration, but merely
remains as an undissolved solid at the bottom of the solution.
The appearance of this extra phase, solid salt, has therefore again
reduced the number of degrees of freedom by one, since the pressure
now depends on a single variable, the temperature. The latter
alters the concentration of the solution in equilibrium with the
solid, and the vapour pressure. When solid salt is present, there
are two equilibria coexisting :

1. Water (vapour) ^± Water (in solution).

2. Salt (solid) ^± Salt (in solution).

In considering equilibrium states of solutions, therefore, an
additional variable enters, which was not involved in the case of
pure substance, viz., the concentration. In a solution of two
substances we need, obviously, only one concentration in order to
define the composition ; the other is then fixed. Thus, if the
.solution contains x per cent, of one component, it must contain
x per cent, of the other. The independent variables,
pressure, temperature, and concentration, are really exactly
analogous to those, pressure, temperature, and specific volume, for

106

INORGANIC CHEMISTRY

CHAP.

a pure substance, since in the latter case we could have taken,
instead of the specific volume, its reciprocal, 1/v, or the density,
which is the analogue of concentration.

Components.— If we have a system composed of phases, some
or all of which are solutions, or pure substances of different
chemical composition not convertible into one another, we shall
require a limited number of substances from which the chemical
composition of every phase may be constituted. The least
number of these substances is called the number of components
of the system. Thus, each of the three phases forming the triple
point of water can be composed of a single substance, water ;
the systems formed from salt and water contain phases all
of which can be built up of one or more of two components, salt
and water.

The phase rule. — Consider the following table, which summarises
results previously described.

Equilibria.

Water (liq.) ^= Water (vap.)

Water (solid) ^± Water (vap.)

Water (solid) :=± Water (liq.)

Water (solid) ;=! Water (liq.)

^Water (vap.)

Salt (dissd.) ^± Salt (solid)

Water (in sol.) z±: Water (vap.)

/ Water (in sol.) ^ Water (vap.)

(Salt (dissd.) ^ Salt (solid)

Number

of
components, C

Number

of
phases, P

Number of
degrees of
freedom, F

In all these cases we remark that a simple relation exists between
the number of phases, of components, and of degrees of freedom,
viz.,

Number of phases -+- Number of degrees of freedom = Number of
components -\- 2.

This relation is perfectly general, and applies to all heterogeneous
systems in equilibrium ; it is called the phase rule (Willard Gibbs,
1876). If we denote the number of phases (p. 7) by P ; the number
of degrees of freedom, or the least number of independently variable
magnitudes (temperature, pressure, and concentrations) which must
be arbitrarily fixed before the state of equilibrium of the system
is completely defined, by F ; and the number of components by C,
then :

P -f F == C + 2.

Examples on the phase rule. — The following examples, to which

vr SOLUTIONS AND THE PHASE RULE 107

the phase rule may be applied, are recapitulated. It will be
seen that the rule is of great value in dealing with solutions.

1. Pure substance ; (7=1.

a. Homogeneous gas, liquid, or solid : P = 1, hence F = I +2—1
= 2. Thus, temperature and pressure, or temperature and
concentration (density), or pressure and concentration, must be
fixed before the state of equilibrium is defined.

b. Phases of a pure substance ; (7 = 1.

(i) Solid^ Liquid, or Solid^ Vapour, or Liquid,"^ Vapour : P = 2,
hence .F = l+2— 2 = 1, i.e., only temperature, or pressure,
or one concentration, can be arbitrarily fixed before the state
of equilibrium is completely defined.

(ii) Solid J± Liquid^ Vapour, i.e., the triple point: P = 3,
hence JP = l+2 — 3 = 0, i.e., no single variable can be
changed without causing complete disappearance of one phase
from the system.

. Solutions, say of two- components ; (7 — 2.

a. Gas ^± Gas (dissd.) : P = 2, hence F = 2 + 2 — 2 = 2, i.e.,
temperature, pressure, or one concentration only can be fixed,
and the system is then in equilibrium. We notice that the phase
rule gives no indication of the way in which the concentration of the
solution depends on the pressure, beyond the fact that it is fixed,
at a given temperature, when the pressure of the gas is fixed.
Henry's law gives a simple proportionality between pressure
and concentration, but this holds only approximately, whereas
the phase rule is quite general, and is not bound by approximate
limitations.

b. Solid^H Solid (dissd.) : P = 2, hence ^ = 24-2-2 = 2, i.e.,
the solubility depends on the temperature and pressure. The
effect of pressure, which is very slight, could have been predicted
by the phase rule.

c. Solid ;=± Solution Z^± Vapour of Solvent : P = 3, hence F = 2 + 2

-3 = 1, i.e., a solution can be in equilibrium with solid and vapour
on)y at a definite pressure (the pressure of the saturated vapour),
and concentration (that of the saturated solution), at a given
temperature.

d. Liquid I z^± Liqmd II, two partially miscible liquids, say ether and
water, existing in two layers in absence of the vapour : P =2,
hence ^ = 2 + 2 — 2=2, i.e., the composition of each layer
is fixed at a given temperature and pressure. The influence of
pressure on the miscibility is small ; it is wholly eliminated
if the vapour phase is present. : P = 3, hence F = 2 + 2— 3 = 1,
i.e., the degree of miscibility depends only on the temperature.

108 INORGANIC CHEMISTRY CHAP.

The eutectic point, the freezing points of solutions, and the effect
of adding iodine to two layers of ether and water, may be
considered by the reader.

The phase rule is seen to be at the same time very simple, and
capable of very extensive application. In the latter, it has led
to the jettisoning of a large bulk of speculative material which
formerly occupied considerable space in the text -books of chemistry.

SUMMARY OF CHAPTER VI

A solution is a homogeneous phase formed from two or more pure
substances, the composition being continuously variable within certain
limits. All states of substances may form solutions.

Henry's law applies to solutions of gases in liquids, and states that
flie solubility is proportional to the pressure. It is an approximate
law only. The solubility of each constituent of a mixture of gases is
proportional to its partial pressure (Dalton's law).

Partition law : if a substance, e.g., iodine, is shaken with two liquids,
e.g., ether and water, which are not, or are only partly, miscible, the,
ratio of the concentrations of the dissolved substance in the two liquid layers
is constant at a given temperature. This ratio is called the partition
coefficient.

The freezing point of a liquid is lowered by a dissolved substance, and
the lowering is proportional to the concentration. This is true only
if pure solid solvent separates on freezing.

The phase rule : the number of components, C, of degrees of freedom,
F, and of phases, P, in a heterogeneous system in equilibrium are
related by the equation P -f- F = C + 2.

EXERCISES ON CHAPTER VI

1. Define : phase, equilibrium, solution, solute, triple point, solu-
bility. Describe a method by which you would determine the solubility
of potassium chlorate in water at various temperatures.

2. State Henry's law. In what way would you proceed to test it in
the case of carbon dioxide ? Define absorption coefficient.

3. From the following data draw the solubility-curves of the salts :
gm./lOOgm. water 0° 10° 20° 40° 60° 80° 100°

(a) Potassium nitrate... 13-3 20-9 32 64 110 169 246

(b) Glauber's salt 5-0 9-0 19-4 49 45 44 42

4. Define partition ratio. The partition ratio for iodine between
carbon disulphide and water is 410 at a given temperature. On
shaking an aqueous solution of iodine with oarbczi disulphide, 35-42
gm. of iodine were found per litre of the disulphide layer. Find the
concentration of iodine in the aqueous layer.

5. State Gibbs's Phase Rule, and explain the terms used. Give four
examples of its application.

vi SOLUTIONS AND THE PHASE RULE 109

6. What is a supersaturated solution ? Describe an experiment
illustrating the production and properties of such a solution.

7. What are cryohydrates, and how are they produced ? What are
they now usually called ?

8. Carbon dioxide is diluted with twice its volume of air, and shaken
with water at 15°. What volume of carbon dioxide should be dissolved
by 1 litre of water ?

9. What experimental evidence would you bring forward in support
of the statement that " sea -water is a solution" ? How can it be
separated into its constituents ? Give three methods which have been
described for separating the constituents of solutions.

CHAPTER VII

THE LAWS OF STOICHIOMETRY

Stoichiometry. — That part of chemistry which deals with the
composition of substances, by weight or volume, is called stoichio-
metry, this word being first used by Jeremias Benjamin Richter,
in his " Anfangsgrunde der Stochiometrie," or " Rudiments
of Stoichiometry " (Breslau, 1792-4), to denote " the art of
measuring the chemical elements," i.e., their combining ratios.

The experimental laws deduced from a study of chemical com-
position are five in number ; four relate to weights and one to
volumes. They are called the Laws of Stoichiometry, or the Laws
of Chemical Combination :

I. The Law of Conservation of Matter, without which there could
be no quantitative investigation of material bodies
(p. 19).

II. The Law of Constant Proportions (Proust, 1799).
III. The Law of Multiple Proportions (Dalton, 1803).
IV. The Law of Equivalents (Richter, 1792), sometimes called the
Law of Reciprocal Proportions, or the Law of Combining
Weights.

V. The Law of Gaseous Volumes (Gay-Lussac, 1808).
The first four laws will be studied in the present chapter ; the
law of volumes is considered in Chapter IX. All the laws have an
experimental basis, and are quite independent of the Atomic
Theory, which, however, gives a simple and rational explanation
of them, as will be seen in the next two chapters.

The law of constant proportions. — This law, asserted by Proust
in 1799, states that : When combination between elements takes place, it
is in definite proportions by weight, so that the composition of a pure
chemical compound is independent of the way in which it is prepared.

If x is the weight of an element X, y the weight of an element

CH. vii THE LAWS OF STOICHIOMETRY 111

V, present in one specimen of a pure chemical compound of X and
y, the ratio x/y is the same in all other specimens of this compound.

" We must," says Proust, " recognise an invisible hand which
holds the balance in the formation of compounds ... a com-
pound is a substance to which Nature assigns fixed ratios, it is, in
short, a being which Nature never creates otherwise than balance
in hand, ponder e et mesura."

This law may appear self-evident ; it was not established, however,
until a long and heated controversy between Proust and Berthollet
had run its course. The latter chemist, a contemporary and
acquaintance of Lavoisier, asserted in his " Chemical Statics "
(1803) that the composition of a compound was variable, and
dependent on its mode of preparation. He relied on the following
experimental evidence :

1. A metal such as lead, when heated in air, absorbs oxygen in
continuously increasing amounts up to a fixed maximum, corresponding
with the formation of red lead, and the colour of the oxide, at first grey,
passes through yellow to red by insensible gradations as the amount of
oxygen increases.

2. A salt formed from a soluble acid and an insoluble base, such as
sulphate of copper, may be precipitated with increasing amounts of a
soluble base, such as potash, to form a continuous series of basic salts,
in which the proportion of acid continuously diminishes. In the case
mentioned, these form greenish-blue precipitates.

3. When mercury is dissolved in nitric acid, it unites with quantities
of oxygen varying continuously from a minimum, when it forms
mercurous salts, to a maximum, when it forms mercuric salts.

4. Aqueous solutions of sulphuric acid, salts, alcohol, etc., and metallic
alloys and amalgams can be formed from their constituents in very
variable proportions.

Proust was able to meet these objections one by one, and overturn
them by experiment.

(1) The members of the supposed continuous series of metallic
oxides were found to be mixtures of two, or a small number, of oxides,
each of definite composition. Thus, the supposed series of oxides of
tin, obtained by calcining the metal in air for varying periods of time,
were all mechanical mixtures of two definite oxides of tin, possibly with
some unchanged metal. These oxides were found by Proust to have
the following compositions :

1. Suboxide of Tin. 2. Protoxide of Tin.

Tin 87 78-4

Oxygen ... 13 21-6

112 INORGANIC CHEMISTRY CHAP.

The " oxide " of composition, tin 80, oxygen 20, for instance, prepared
by Berthollet, was a mixture of 81-4 parts of protoxide with 18-6 of
suboxide.

(2) The supposed basic salts of copper of variable composition were
all hydrated oxide of copper, imperfectly freed from sulphate by washing.

(3) Mercury on dissolving in nitric acid forms only two salts :
mercurous nitrate, formed when excess of metal is treated with cold
dilute nitric acid, and mercuric nitrate, which is produced from the
metal and excess of hot concentrated nitric acid. The other supposed
salts were mixtures of these.

Berthollet was forced to recognise that in many cases substances
of definite composition could be formed, but he regarded these as
exceptional. In them the particular proportions of the elements
gave the compound which was least soluble, or most volatile, or
densest. Thus, " it so happens that salts separate out by crystal-
lisation in the neutral state, because in the neutral state the in-
solubility is greatest."

The fifth class mentioned under Berthollet 's evidence gave Proust
a good deal of trouble. He replied by pointing out the difference
between a pure substance and a solution (or mixture). He says :

" Is the power which makes a metal dissolve in sulphur different from
that which makes one sulphide dissolve in another ? I shall be in no
hurry to answer this question, legitimate though it be, for fear of losing
myself in a region not sufficiently lighted up by the facts of science ; but
my distinctions will, I hope, be appreciated all the same when I say :
The attraction which causes sugar to dissolve in water may or may not
be the same as that which makes a fixed quantity of carbon and of
hydrogen dissolve in another quantity of oxygen to form the sugar of
plants, but what we do clearly perceive is that these two kinds of
attraction are so different in their results that it is impossible to con-
found them."

Unfortunately, the matter is not always so simple ; the alloys
formed from mixtures of metals are sometimes simply mixtures of
the metals, each of which has crystallised out separately on cooling ;
sometimes they are homogeneous solutions, and sometimes they
contain definite compounds, of the metals. It is only comparatively
recently that it has been possible to decide to which class a par-
ticular alloy belongs (Chapt. XXXVII). Proust was, therefore,
wise in refusing to be in a hurry to answer this question.

The exactness of the law of constant proportions was established
by the experiments of Stas (1865) ; Marignac (1860) had previously
suggested that very slight differences might occur in the com-
positions of compounds made in different ways, but Stas's work

THE LAWS OF STOICHIOMETRY

113

showed that, if such differences existed, they did not exceed 1 part
hi KM), 000, and were within the limits of experimental error.

/SO///VT of ammonia.

1 Ammonium sulphate

2 Potassium nitrite
and zinc ...

I) Ammonium sulphate

Do.

5 Ditto second crop of
crystals, sublimed
in pure ammonia
gas and dried over
sulphuric acid

G Same specimen of
ammonium sul-
phate, but dried in
hydrogen gas and
ammonia gas at
180°

Source of hydro -
bromic acid.

Source of silver

Potassium bromide! Reduced with
sodium formate !

Do.

Pure bromine

Do.

Ammon.

bromide

reacting

with 107-93

of silver.

97-996
97-989

Do.

98-001
97-990

Reduced with
cuprous - am-
mon. sulphite

Reduced with
formate
(a) unfused ...
(6) fused on
pure calcium
phosphate ...

Reduced with
formate and
fused on pure
calcium phos-
phate

Same specimen as
in (5)

Same specimen
as in (5)

97-999

97-994
98-003

97-997

98-000

Silver bromide from 4 (a) reduced in hydrogen and the
Silver fused (a) on calcium phosphate with hydrogen ... 97-995
(6) on pure lime with hydrogen (did not appear !
perfectly bright and pure) 97-984

As an example of the law of constant proportions, the above results of
Scott (1901) on the analysis of ammonium bromide may be

114 INORGANIC CHEMISTRY CHAP.

quoted. Stas had been unable to obtain a specimen of this salt which
remained perfectly white on heating ; this was effected by Scott, who
used perfectly pure ammonia and hydrobromic acid, prepared in
different ways. The salt was then precipitated with silver nitrate made
from different specimens of silver.

Isomerism. — The law of constant proportions asserts that a
definite compound has a fixed chemical composition. The con-
verse is not true : the same elements, combined in the same pro-
portions by weight, may form two or more different substances,
with characteristic physical and chemical properties. This property
is known as isomerism, and the different substances of the same
composition are called isomers. Chemical composition alone does
not uniquely determine a pure substance.

Thus, red mercuric iodide, on heating to 126°, changes into a yellow
form, of identical composition. This remains yellow on cooling, but
changes into the red form when rubbed.

An element may also exist in various forms, which are called
allotropic modifications, or allotropes. Allotropy is one form of iso-
merism.

Thus, sulphur, on heating, melts to a clear, mobile, pale yellow liquid.
On further heating this is suddenly transformed into an orange -yellow
viscous mass, which darkens as heating is continued, until at 440° it
is almost black. The liquid is then less viscous, and if poured into cold
water forms a brownish-yellow, transparent, sticky and elastic mass.
On standing for a few days, this slowly becomes opaque, lemon-yellow,
and brittle ; it is reconverted into ordinary sulphur. Equal weights
of crystalline and plastic sulphur, when burnt in oxygen, yield equal
weights of the same substance, sulphur dioxide. They both consist of
the same element, sulphur.

Isotopes. — The unique composition of a pure substance has come
to be regarded as a self-evident fact. Soddy and Hyman
(1914) found, however., that specimens of lead chloride, prepared
respectively from thorium and uranium minerals containing lead,
differed in composition by I part in 225, although they were identical
in chemical properties. This startling result was confirmed by
Richards and Lembert (1914). It appears that there are different
varieties of lead, which combine in different proportions with
chlorine. These different varieties of an element, which appear to be
identical in chemical properties but may have different combining
proportions, are called isotopes. Their existence, which extends to
other elements besides lead, makes the question of the combining
ratios of elements, and the definition of an element, much more
difficult than was formerly supposed (see Chapter LI).

VIT THE LAWS OF STOICHIOMETRY 115

The Law of Multiple Proportions. — As a result of some theoretical
speculations on the atomic constitution of matter, John Dalton,
some time between 1802 and 1804, and probably in 1803, was led
to assume that : If two elements combine to form more than one com-
pound, the weights of one element which unite with identical weights of the
other are in simple multiple proportion.

Although Proust was acquainted with different oxides of each of
the metals, tin, copper, and iron, his analyses were not sufficiently
accurate to disclose any simple relation between the weights of
oxygen combined with identical weights of metal or vice versa.
Thus, in the two oxides of tin (p. Ill), the weights of tin combining
with 100 parts of oxygen are in the ratio 1 : 1-87. According to
Dalton's ideas, the ratio should be exactly 1 : 2, and he made further
analyses to confirm this. Dalton's analyses were no more exact than
the former, but those subsequently made by Berzelius established
the accuracy of the law in question.

Dalton, by mixing 100 vols. of air with 36 vols. of nitric oxide
over water in a narrow tube (5 in. X 0-3 in.), obtained a residue of
80 vols. of nitrogen after all the oxygen of the air had combined with
the nitric oxide to form red fumes, which were absorbed by the
water. But if the experiment was performed in a wide cylinder,
72 vols., i.e., 36 x 2 vols., of nitric oxide could be added, 80 vols.
of nitrogen again remaining. Thus, " . . . oxygen can combine
with a certain portion of nitrous gas, or with twice that portion, but
with no intermediate quantity."

Analyses of two oxides of nitrogen by Davy, and of two hydrides
of carbon by Dalton (1804), the latter results probably rounded off,
also confirmed the law : .

Nitric oxide. Nitrous oxide.

Nitrogen . . . . 79-8 164-8 = 79-8 x 2-00

Oxygen . . . . 100 100

Marsh gas. Olefiant gas.

Carbon . . . . 4*3 4'3

Hydrogen .... 2 1

The most striking example of the law of multiple proportions is
furnished by the five oxides of nitrogen. The percentage compositions
by weight of these five compounds are as follows : —

Nitrous

Nitric

Nitrous

Nitrogen

Nitric

oxide.

anhydride.

dioxide.

anhydride.

Nitrogen . .

63-7

40-7

36-9

30-5

25-9

Oxygen . .

36-3

53-3

63-1

69-5

74-1

i 2

116 INORGANIC CHEMISTRY CHAP.

The weights of oxygen combined with 100 parts of nitrogen in
these compounds are found by proportion, and are as follows :

57 114 171 228 285

If all these numbers are divided by the least, 57, we obtain the
series :

12345

Thus, the weights of oxygen combining with identical weights,
100 parts, of nitrogen to form the five compounds are in the simple
proportion 1:2:3:4:5.

EXPT. 55. — Weigh out two portions of 6-35 gm. of iodine. Add
one in small quantities at a time to 10 gm. of mercury in a small mortar,
triturating the contents after each addition of iodine, and adding one or
two drops of alcohol. The mixture of 10 gm. of mercury and 6-3 gm. of
iodine is converted into a green powder (mercurous iodide). To this
add a further 6-3 gm. of iodine and a few drops of
alcohol, and triturate. The 10 gm. of mercury and
12-7 gm. of iodine give a red powder (mercuric iodide).
Under the microscope, these two substances are seen
to be homogeneous. Thus, in mercuric iodide the
same weight of mercury is combined with twice the
• amount of iodine contained in mercurous iodide.

EXPT. 56. — Wrap 0-5 gm. of bicarbonate of potash
in tissue paper and pass it to the top of a graduated
tube filled with mercury, the upper part containing
5 c.c. of concentrated hydrochloric acid (Fig. 71).

FIG. 71. — Experi- Carbon dioxide is evolved. Heat 1 gm. of bicarbonate
meat on Multiple
Proportions. in a platinum crucible for a lew minutes : it loses

part of its carbon dioxide, forming carbonate of
potash. If this is treated with acid, it evolves exactly the same
volume of gas as the 0-5 gm. of bicarbonate. Hence the bicarbonate,
on heating, loses exactly half its carbon dioxide in forming the carbonate.

Experiment 56 is due to William Hyde Wollaston (1808) ; in
the same year Thomas Thomson showed that oxalic acid reacts
with potash in two proportions, producing a neutral and an acid
salt, and the acid oxalate requires, for identical weights of potash,
exactly twice as much acid as the neutral salt. Wollaston discovered
a third oxalate, and found the weights of oxalic acid reacting to be
in the proportion 1 : 2 : 4. The law of multiple proportions therefore
applies not only to elements, but also to compounds which interact
chemically.

The exactness of the law of multiple proportions is well illustrated
by the results of Stas (1849) and others on the composition of the

vii THE LAWS OF STO1CHIOMETRY 117

two oxides of carbon, carbon monoxide and carbon dioxide. Carbon
dioxide was prepared by passing oxygen over a weighed amount of
pure charcoal, diamond, or graphite, heated in a tube, and absorbing
the gas in tubes containing caustic potash. Carbon monoxide was
also oxidised to dioxide by passing it over red-hot copper oxide :
carbon monoxide -f- copper oxide — carbon dioxide -4- copper.

One hundred parts of carbon dioxide were found to contain 27-278
parts of carbon. The weight of carbon monoxide yielding 100 parts
of carbon dioxide was 63-640. Thus :

One hundred parts of carbon dioxide are produced from :

63-640 parts of carbon monoxide and 100 — 63-640 = 36-360

parts of oxygen.

27-278 parts of carbon and 100 — 27-278 = 72-722 parts of oxygen.

Again, 63-640 parts of carbon monoxide contain 27-278 parts of carbon

and 63-640 - 27-278 = 36*362 parts of oxygen. Thus 27-278 parts of

carbon are combined in carbon monoxide with 36-362 parts of oxygen,

and in carbon dioxide with 72-722 parts of oxygen.

But 36-362 : 72-722 : : 1 : 1-99995, which differs from the exact ratio
1 : 2 by only 1 part in 40,000, i.e., within the errors of experiment.

The law of equivalent proportions. — In 1766 Cavendish called a
given weight of potash the equivalent of a (different) weight of lime
when both neutralised identical weights of an acid. In 1788 he
showed that the quantities of nitric and sulphuric acids which
neutralised two identical weights of potash would also neutralise two
identical weights of marble, different from those of the potash.
This was the first clear recognition of equivalent weights of sub-
stances which interact chemically.

Experiments on the compositions of salts, and the proportions in
which they interact chemically, were made by C. F. Wenzel, and
published in his " Lehre von der Verwandtschaft der Korper "
("Theory of the Affinity of Bodies"), Dresden, 1777.* Wenzel
was credited by Berzelius, apparently by an oversight, with the
discovery of the law of equivalents. This is not confirmed by an
examination of the book, which is written in an involved and obscure
style.

In one experiment. Wenzel discusses the reaction between silver
chloride and mercury sulphide, producing silver sulphide and mercuric
chloride. He found that ^ oz. of luna cornea (silver chloride) contained
180t°r7 grains of silver. From an analysis of silver sulphide (" geschwe-
feltes Silber "), he found that this 180T9^ grains of silver are combined
with 26| grains of sulphur. An analysis of cinnabar (mercuric sulphide)
showed that 26| grains of sulphur form 125| grains of cinnabar.

* There is a copy of this very rare book in the British Museum Library.

118

INORGANIC CHEMISTRY

CHAP.

Now £ oz. of luna cornea contains 53T7g grains of " Salzsaure "
(really chlorine), but by subtraction of the silver, ]80^ grains, from 240
grains, this amount would be 59y7g grains, instead of 53j7ff. An
analysis of corrosive sublimate (mercuric chloride) showed that 53T7ff
grains of acid require 159| grains of mercury, and, from the analysis of
cinnabar, this would correspond with 202^ grains of cinnabar, instead
of 125^, as found in the first set of analyses.

Wenzel therefore remarks that : "125^ grains of cinnabar would not
separate all the acid in the luna cornea" Further, if the mixture of
cinnabar and luna cornea be sublimed, " the acid of the * Hornsilber'
(luna cornea} rises with the mercury out of 202^ grains of cinnabar as a
corrosive sublimate ; the silver, on the other hand, remains combined
with only so much sulphur as is contained in 125^ grains of cinnabar."
The inference is that the excess of sulphur remains uncombined. In
other cases, Wenzel actually refers to uncombined residues from double
decompositions, and suggests that they be used up by adding other
substances. It therefore seems wide of the mark to suggest that
Wenzel had any idea of the law of equivalents, or that his analyses were
more exact than those of his contemporaries.

The generalisation of Cavendish's experiments is due to J. B.
Richter, whose results are contained in his treatise on stoichiometry
(see p. 110), 1792-4. Richter's reasoning is quite unnecessarily
obscured by his attempts to derive mathematical relationships which
do not exist ; stripped of its verbiage, and exhibited in all its
essentials, it appears in the German translation, by G. E. Fischer,
of Berthollet's " Researches on the Laws of Affinity " (1802). In
this the first table of equivalent weights of acids and bases is given, a
portion of which is reproduced below.

Bases.

Alumina

Ammonia

Lime

Soda

Potash

Baryta

525 Fluoric ...

672 Carbonic

793 Muriatic

859 Oxalic ...

1605 Sulphuric

2222 Nitric ...

Acids.

427

577

712

755

1000

1405

" The meaning of this table," said Fischer, " is that, if a substance
is taken from one of the two columns, say potash from the first,
to which corresponds the number 1605, the numbers in the other
column indicate the quantity of each acid necessary to neutralise
these 1605 parts of potash. There will in this case be required 712
parts of muriatic [hydrochloric] acid, 577 parts of carbonic acid, etc.
If a substance is taken from the second column, the first column is
to be used to ascertain how much of an earth or of an alkali is required
to neutralise it."

This table of twelve numbers enables us to calculate, by addition

vii THE LAWS OF STOICHIOMETRY 119

in pairs, the composition of thirty-six salts. By the analysis of six
of the latter, say those corresponding with the constituents on the
horizontal lines (e.g., sulphate of potash), the compositions of the
remaining thirty may be found.

Richter's result is a special case of the law of equivalent pro-
portions : the weights of two (or more) substances which separately
react chemically with identical weights of a third are also the weights which
react with each other, or simple multiples of them.

An important case of the law is that which applies to the com-
bination of elements. The combining weights, or equivalent weights,
or simply equivalents, of the elements, are really the most fundamental
values, since the equivalent weights of compounds are formed
additively from those of their constituent elements.

Equivalents of the elements. — It is found that 23 gm. of sodium com-
bine with 1 gm. of hydrogen to form sodium hydride ; 35-2 gm. of
chlorine combine with 1 gm. of hydrogen to form hydrogen chloride.
The equivalent weights of sodium and chlorine, with respect to
combination with hydrogen, are therefore 23 and 35-2, respectively.
Now sodium and chlorine also combine together to form sodium
chloride, and it is found that 23 parts of sodium combine with 35*2
parts of chlorine to form sodium chloride.

Thus, the weights of sodium and chlorine which separately combine
with 1 part by weight of hydrogen are

the weights in which these two elements HYDROGEN

combine with each other. This fact
may be illustrated by the annexed
diagram.

If 23 gm. of sodium are heated in
hydrogen chloride gas, it is found that
1 gm. of hydrogen is displaced, whilst 35:2
35-2 gm. of chlorine combine with the CHLORINE SODIUM

23 gm. of sodium to form sodium

chloride. Thus, 23 parts of sodium can combine with 1 part of
hydrogen, and can also displace it from its combination with
another element.

The equivalent of an element is defined as that weight of it which combines
with, or displaces, 1 part by weight of hydrogen.

Hydrogen is taken as the standard element for reasons of sim-
plicity, because it is found that no element has an equivalent less
than that of hydrogen.

The conception of an equivalent implies that, when once the
equivalent of a single element has been determined with respect to
hydrogen, the equivalent of that element may be used instead of

120 INORGANIC CHEMISTRY CHAP.

hydrogen in the determination of other equivalent weights. Thus,
having found that the equivalent of chlorine with respect to hydrogen
is 35-2, we may use 35-2 parts of chlorine instead of 1 part of
hydrogen in finding the equivalent of an element which combines
with chlorine but does not combine with hydrogen. In the case
of sodium, which combines with both hydrogen and chlorine, the
equivalents are found to be identical. In other cases, the element
may displace hydrogen but does not combine with it ; e.g., zinc,
which evolves hydrogen from dilute acids, does not form a hydride.
It is again found that the weight of such an element which com-
bines with 35-2 parts of chlorine displaces 1 part of hydrogen.
In the case of elements which neither combine with nor displace
hydrogen, such as gold, the equivalent weight may be determined
with respect to combination with chlorine, and is thus fixed in an
indirect manner. The equivalents of such elements are then simply
the weights which combine with or displace equivalent weights of
other elements, which have been ascertained directly with respect
to hydrogen.

The equivalent of oxygen may be calculated from the composition
of water. In this way (p. 64) it is found to be 7-94. This number
is of importance, since the equivalents of metals are sometimes
determined with respect to oxygen, by converting the metal into
the oxide. In this case, the equivalent is the weight combining
with 7-94 par^s of oxygen.

The equivalent of a compound is the sum of the equivalents of its
constituent elements.

The determination of equivalents. — Equivalents are determined
experimentally in various ways.

(1) The weight of the element combining with or displacing 1 part
of hydrogen is found. This is applicable to metals which dissolve in
acids, or alkalies, with evolution of hydrogen, the volume of which is
measured.

(2) The weight of metal displaced from a solution of one of its salts by
the equivalent of another metal, falling in class ( 1 ) is found. Thus, the
equivalent of zinc is found by the measurement of the hydrogen
evolved by zinc from an acid, and the equivalent of copper is then
determined by weighing the copper displaced by the equivalent weight
of zinc from a solution of copper sulphate.

(3) The weight of the element combining with 7-94 parts of oxygen
is found ; the combination may take place directly, as when magnesium
is heated in air or oxygen, or indirectly, as when tin or copper is treated
with nitric acid, and the product heated to redness. If the oxygen
compound is decomposed on heating, e.g., mercuric oxide, or potassium
chlorate, the weight of oxygen liberated is found, and the equivalent
of mercury, or of potassium chloride, e.g., so determined.

VIJ

THE LAWS OF STOICHIOMETRY

121

(4) The weight of silver, the equivalent of which has been determined
directly with respect to chlorine, required to precipitate a known weight
of the chloride of an element, e.g., potassium chloride, gives the equiva-
lent of the latter.

(5) A given weight of one compound, composed of elements of
known equivalents, may be converted into another compound, con-
taining the element of which the equivalent is desired. Thus,
potassium chloride is converted into potassium nitrate to determine the
equivalent of nitrogen (those of potassium, chlorine, and oxygen being
known.)

Since elements sometimes combine in more than one proportion,
it follows that an element may have more than one equivalent.
The law of multiple proportions then shows that the different equiva-
lents of an element are related in simple multiples. Thus, carbon
forms two compounds with oxygen, containing, for 7-94 parts of
oxygen, 2-978 and 2-978 x 2 parts of carbon, respectively (p. 117).

EXPERIMENTS ON EQUIVALENTS

EXPT. 57. — Weigh out about 1 gm. of pure zinc into the small tube,
A (Fig. 72). Lower this carefully into the dilute sulphuric acid
(1:5 by vol.) in the flask,
B (to which two or three
drops of copper sulphate
solution have been added),
and fit the flask with the
rubber stoppers, C and C",
to the bottle, D, containing
about 1000 c.c. of water,
previously filling the tube E
with water and closing the
pinchcock, F. When the
flask, B, is in position open F.
If the stoppers are tight,
only a little water should run
into the graduated cylinder,
G, and this is poured out.
Shake the flask, B, so as to bring the zinc into the acid. When the
evolution of gas ceases, allow the apparatus to stand till the tem-
perature is uniform ; then, by raising or lowering G, bring the water
levels to equality in G and D. Close the clip, F, remove the cylinder,
and find the volume of gas evolved. Take the temperature of the
water, read the barometer, and find the volume of hydrogen reduced to
S.T.P. One c.c. of hydrogen at S.T.P. weighs 0-00009 gm. Find
the weight of zinc which displaces 1 gm. of hydrogen, i.e., the equivalent

FIG. 72. — Apparatus for Determination of
Equivalents.

12i> INORGANIC CHEMISTRY OHAF,

of zinc. Repeat with iron (pure \viiv). magnesium (use very dilute
acid), and aluminium (use strong caustic soda in B, and warm, if
neeessary). In each cast1 find the equivalent of the metal.

EXPT. 58. — Weigh out 1-2 gm. of pure zinc and place it in a beaker
containing a solution of copper sulphate, to which one or two drops of
dilute sulphuric acid have been added. A red spongy deposit
of copper is produced. When all the zinc has disappeared, filter
through a weighed paper, wash the copper with hot water till the filtrate
no longer becomes turbid with barium chloride (p. 12), dry in an air
oven at 120°, and weigh. From the result of EXPT. 57, find the
equivalent of copper. Repeat with magnesium and iron instead of
zinc. From the value for copper found with zinc, calculate the
equivalents of magnesium and iron, and compare with those found in
EXPT. 57.

EXPT. 59. — Weigh out about 0-5 gm. of magnesium ribbon into r,
porcelain crucible with lid. Heat over a Bunsen burner until the metal
has burnt into oxide, then remove the lid, and continue heating for ten
minutes. Cool in a desiccator and weigh. Calculate the equivalent of
magnesium, and compare with that previously found.

EXPT. 60. — Weigh out 1-2 gm. of pure tinfoil into a porcelain dish
with a watch-glass cover. Add a few drops of concentrated nitric acid
and replace the glass. Repeat till the action ceases, then heat carefully
on a sand-bath till the excess of acid is driven off. Wash the oxide from
the glass into the dish with a wash-bottle, and evaporate to dryness
without cover, carefully avoiding spirting. Heat strongly for five
minutes, cool in a desiccator, and weigh. Find the equivalent of
tin (c/. p. 26). Repeat the experiment with copper.

Exact determination of equivalents. — The determination of the
equivalents of a limited number of elements with all possible exact-
ness was the life-work of the Belgian chemist J. S. Stas (1813-1891),
whose numbers for carbon, nitrogen, sulphur, chlorine, bromine,
iodine, lithium, sodium, potassium, lead, and silver were accepted
for a number of years as the most accurate values, and regarded
with almost superstitious reverence.

Stas began his researches with an analysis of potassium chlorate,
which on heating gives off oxygen and leaves potassium chloride. Since
the ratio hydrogen/oxygen was not certainly known at that time, Stas
proposed as the basis of his numbers the equivalent of oxygen, which
he took, not as 1, but as 8-00. He found that 127-2125 gm. of potassium
chlorate gave on heating 77-4023 gm. of potassium chloride, hence the
oxygen given off weighed 49-8102 gm. Potassium chlorate is known
to contain 6 equivalents of oxygen, hence the equivalent of potassium
chloride, xy is given by :

6 X 8 : x ; ; 49 8102 : 77-4023, ,% v

vn THE LAWS OF STOICHIOMETRY J±{

14-427 gm. of potassium chloride gave, on precipitation with silver
nitrate solution, 27-733 gm. of silver chloride, hence the equivalent of
silver chloride is given by the proportion :

74-59 : x : : 14-427 : 27-732, .'. x = 143'37.

101-519 gm. of pure silver when burnt in a current of chlorine gave
134-861 gm. of silver chloride, so that the equivalent of silver is given
by the proportion :

143-37 : x : : 134-861 : 101-519, .'. x = 107'93.

Hence the equivalent of chlorine is 143-37 — 107-93 = 35*44, and
the equivalent of potassium is 74-59 — 35-44 — 39*15.

The results of Stas may be summarised as follows :

Oxygen 8 -00 (standard) Silver 107-93

Chlorine 35-44 Potassium 39 • 1 5.

In 1895 Morley determined the ratio hydrogen /oxygen with great
care and found 1 : 7-939 ; Scott (1893), and Burt and Edgar (1915),
in most accurate researches, found 1 :7-938. The equivalent of
chlorine was determined directly (Dixon and Edgar, 1905, and
Edgar, 1908), by the combustion of the gas in hydrogen, and
weighing the hydrochloric acid, to be 35-186, with reference to
hydrogen as unity. This number was exactly confirmed by [a
determination of the density of hydrochloric acid gas (Gray and
Burt, 1909), and the decomposition of the latter by heated alu-
minium, with liberation of hydrogen. The equivalent of chlorine
referred to oxygen = 8-00 is thus

35-186 x 8-00 4- 7-945 = 35463,

which differs from Stas's figure by as much as 1 in 1500.

This large discrepancy, confirmed by newer determinations,
led to a suspicion that some at least of Stas's figures must be
affected by systematic errors, and this was found to be the case.
Even carefully recrystallised potassium chlorate always contains
potassium chloride, and silver chloride, when precipitated from a
solution of potassium chloride, always carries down some of the
latter salt, which cannot be removed by washing.

The equivalents of the majority of the important elements are based
on the equivalent of silver. Oxygen appears in few direct ratios.
Since the equivalent of silver may be referred to that of hydrogen
through the single intervening link of chlorine, i.e., from the two ratios :
silver /chlorine, and chlorine /hydrogen, both of which are very accurately
known, hydrogen could more conveniently be adopted as a
practical standard than oxygen. Its theoretical advantages are numerous.
The fundamental derived unit, silver, cannot be referred directly to
oxygen : on the oxygen standard, the value for nitrogen is involved as

124 INORGANIC CHEMISTRY CH. vn

an intermediate link, since the ratio sliver / silver nitrate is determined.
But the most accurate value for nitrogen was derived from the analysis
of ammonia, a hydrogen compound.

The single practical advantage of equivalents referred to oxygen
= 8 is that in the case of a few common elements the numbers are
more nearly whole numbers than is the case with hydrogen equivalents
(p. 145).

SUMMARY OF CHAPTER VII

The quantitative laws of chemistry relating to weight (or mass) are :
(1) The Law of Conservation of Matter (Chapter II); (2) The Law
of Constant Proportions (Proust, 1799) : when combination between
elements occurs, it is in definite proportions by weight ; (3) The Law of
Multiple Proportions (Dalton, 1803) : when two elements form more than
one compound, the weights of one element which combine with identical
iveights of the other are in simple multiple proportion ; (4) The Law of
Equivalent Proportions (Richter, 1792) : the weights of two substances
(e.g., elements) which separately react (e.g., combine) with identical
weights of a third, are also the weights in which they react with each other,
or simple multiples of them.

The equivalent of an element is primarily defined as the weight which
combines with, or displaces, 1 part by weight of hydrogen. The relation
of equivalence may then be extended throughout the whole series of
elements, including those which do not react with hydrogen. An ele-
ment may have more than one equivalent : the taw of multiple propor-
tions then shows that the equivalents must be related as whole numbers,
usually small.

EXERCISES ON CHAPTER VII

1. What are the laws of stoichiometry ? State those relating to
weight.

2. Give a short account of the nature and results of the controversy
between Proust and Berthollet. What difficulty had Proust to
explain after his work, and what account did he .give of it ?

3. Describe experiments to illustrate the laws of constant, multiple,
and equivalent proportions. What is known as to the degree of accuracy
of these laws ?

4. 3-3665 gm. of zinc displaced 1212-09 c.c. of hydrogen, measured at
747-84 mm. and 10-73°, from dilute sulphuric acid. Calculate the
equivalent of the metal.

5. 150-000 gm. of silver heated in sulphur vapour gave 172-2765
gm. of silver sulphide. 81-023 gm. of silver sulphate on reduction in
hydrogen gave 56-071 gm. of silver. Assuming that the ratio
silver I sulphur is the same in both compounds, and that silver sulphate
contains 4 equivalents of oxygen, find the equivalents of silver and
sulphur (Oxygen = 8).

CHAPTER VIII

THE ATOMIC THEORY

Atoms. — The simplicity of the laws of chemical combination, or of
stoichiometry, considered in the preceding chapter, leads inevitably
to the conviction that they have their counterpart in some simple
character of matter. The explanation of these laws presupposes
some idea of the structure of matter. This may, no doubt, lie beyond
the possibility of experimental apprehension, even when the observer
is assisted by the most powerful of microscopes, and to this extent
will remain a hypothesis, or a guess as to a possible cause. We shall
see later, however, that an increase in the power of the ultra-micro-
scope a hundred- or even ten-fold would bring us within reach of
the direct perception of the underlying structure of material bodies.

Two possible guesses as to the ultimate structure of matter at once
present themselves, and in fact originated in those distant times
when history as we know it had scarcely begun. The first hypo-
thesis sees matter as a continuous structure, completely filling the
space occupied by bodies in the same way as jelly fills a mould.
The second hypothesis sees matter filling space discontinuous^,
with interstitial gaps, much as small shot fills a barrel. The first
view is associated with the Eleatic school, founded by Xenophanes
(B.C. 576-480) ; the second is so much older that it is impossible to
say where it originated. According to some it arose in India, about
1200 B.C., and passed to Greece, where it was taught by the old
philosophers. Others think it originated with the Greeks them-
selves. This was the atomic hypothesis, which postulated the division
of matter into exceedingly small particles, or atoms, incapable of
further division by physical means.

Speculations on the atomic hypothesis occupied the Greek
philosophers Anaxagoras, Leukippos, and Demokritus. Van Hel-
mont, Lemery (1675), Boerhaave (1724), Boyle, and Newton
(1642-1727) made more scientific use of the hypothesis ; the last,
although the author of the dictum hypotheses non fingo, was a
thoroughgoing atomist. Newton gave a mathematical demon-
stration of Boyle's law on the basis of the hypothesis that gases

125

126 INORGANIC CHEMISTRY CHAP.

consist of atoms repelling one another with forces inversely pro-
portional to the distances. Boscovitch also made extensive appli-
cation of a similar theory, but considered the atoms as mere points,
or centres of force, endowed with mass.

Bryan, and William Higgins, in 1777 and 1789 respectively, made
some applications of Newton's atomic theory to chemistry, but the
merit of having independently elaborated a chemical atomic theory

JOHN DALTON.

which was capable of co-ordinating all the known facts, and of being
modified and extended with the progress of the science, belongs
unquestionably to John Dalton (1766-1844).

John Dalton was born at Eaglesfield, a village near Cockermouth in
Cumberland, and was throughout his life solely dependent upon his own
exertions. As a boy he earned a living partly by teaching the rustic
youth, and partly as a farm labourer. In 1781 he met Mr. John Gough,

vin THE ATOMIC THEORY 127

the blind philosopher of Kendal, whose influence on his life Dalton
often gratefully recognised. After a period of study with Gough,
including the writings of Newton, Dalton removed to Manchester,
where the rest of his life was spent in scientific teaching and research.
Dalton's manuscript note-books were discovered in the Archives of the
Literary and Philosophical Society of that city by Roscoe and Harden
in 1895, and from them it has been possible to trace, though imperfectly,
the origin and development of Dalton's atomic theory. It unquestion-
ably arose from the influence of Newton.

The origin of Dalton's atomic theory. — Apart from the influence
of Newton, it is difficult to say what led Dalton to his atomic theory.
Meldrum (1910) has shown that Dalton himself gave, at various
times, four different accounts of its origin : (1) That communicated
to his biographer, William Charles Henry, which attributed the
theory to the influence of Richter, may be dismissed, since Richter
is not mentioned in Dalton's note-books until 1807, whereas the
atomic theory was certainly in existence in 1803, and probably in
1801-2. (2) Thomas Thomson's account, written after an interview
with Dalton in 1804, was generally accepted before Roscoe and Har-
den's publication of the new facts, and attributed the origin of the
theory to Dalton's attempt to explain the Law of Multiple Pro-
portions, as exemplified by his discovery of the composition of marsh
gas and of ethylene (p. 115). But the analyses of these gases were
not made until 1804, whereas the first list of atomic weights was
published in 1803. (3) Dalton's notes of lectures, given at the
Royal Institution in 1810, trace the theory to some speculations on
" mixed gases " (i.e., on the law of partial pressures), made in 1801-2,
and this is accepted by Roscoe and Harden, since it is the only
account agreeing with the dates, and with the fourth source of
information, viz. : (4) Dalton's manuscript note-books, preserved
in Manchester.

It seems very probable that Dalton was led to his theory on purely
physical lines ; it preceded the law of multiple proportions, and the
latter was deduced from it. Dalton's experiments on multiple
proportions appear to have been confirmatory only.

Dalton's atomic theory. — The atomic theory of Dalton, the great
guiding principle of the whole of modern chemistry, is so simple
that, as Lothar Meyer has said, " at first sight it is not illuminating."
It asserts that :

(1) The chemical elements are composed of very minute particles
of matter, called atoms, which preserve their individuality in all
chemical changes.

Dalton was firmly convinced that these atoms are indivisible ; he
was wont to say : " Thou knowest thou canst not cut an atom," and

128 INORGANIC CHEMISTRY CHAP.

when referred to the sesquioxides (p. 134), which apparently contain H
atoms of oxygen to 1 atom of the other element, replied : " but they
are two to three, " i.e., 2 atoms of element to 3 atoms of oxygen.

(2) All the atoms of the same element are identical in all respects,
particularly in weight. Different elements have atoms differing
in weight. Each element is characterised by the weight of its
atom.

The absolute weights of atoms, as Dal ton realised, are very small
indeed, and he therefore directed his attention to the determination
of the relative weights, taking the weight of the lightest atom, that
of hydrogen, as unity. The atomic weight of an element is then the
number giving the ratio of the weight (or mass) of an atom of that
element to the weight (or mass) of an atom of hydrogen.

If the absolute weight (or mass) of any one atom can be deter-
mined, those of all the others are found by simple multiplication
of this by the ratios of the atomic weights. In recent years the mass
of the hydrogen atom has been found by several different methods,
giving results in surprising agreement (cf. p. 268). It is 1 -66 X 10 ~24
gm. Thus, 1 c.c. of hydrogen, at S.T.P., weighing 0-00009 gm.,
contains 54 X 1019 atoms. The weight of the heaviest atom
known, that of uranium (atomic weight 236), is

236 X 1-66 x 10~24 = 3-92 x 10~22 gm.

(3) Chemical combination occurs by the union of the atoms of
the elements in simple numerical ratios, e.g., 1 atom A -f- 1 atom B ;
1 atom A +2 atoms B ; 2 atoms .4 + 1 atom B ; 2 atoms A + 3
atoms B, etc.

The aggregate of two or more atoms in a compound, called a
" compound atom " by Dalton, is now named a molecule (i.e.,
" a small mass "). A chemical compound contains its elements,
since the atoms of these are present in the molecule, and may be
recovered in the form of the original elements by decomposition.
If mercury, for instance, is converted into the red oxide by heating
in air, and the oxide then decomposed at a higher temperature, the
same mercury is recovered as was used in the synthesis of the oxide.

Deduction of the laws of stoichiometry.— The empirical laws
discussed in the last chapter follow as almost obvious consequences
of the atomic theory.

(1) Since the atoms are indestructible in chemical changes, they
preserve their masses in all such changes, and the mass of a com-
pound is the sum of the masses of its elements. This is the Law
of Conservation of Mass, or of Matter (p. 19).

(2) Just as all the atoms of an element are alike in all respects,

viii THE ATOMIC THEORY 129

so also the molecules of a compound are identical, and are com-
posed of the same number of atoms of the same elements. Thus,
a compound has a unique composition. This is the Law of Constant
Proportions (p. 110).

(3) If two elements combine in more than one proportion, the
molecule of one compound must be formed by adding a whole
number of atoms of one or both elements to one or more molecules
of the other compound. This is the Law of Multiple Proportions.

(4) Compounds of the elements A and C must be formed
according to the scheme : m atoms A -f- n atoms C. Compounds
of the elements B and C must be composed of : x atoms B -f y
atoms C. Compounds of the elements A and B must contain :
p atoms A + q atoms B. But x, y, m, n, p, q are whole numbers,
usually small. Hence p, q are either the same as m, x, or whole
multiples of them, usually small. This is the Law of Equivalent
Proportions.

The equivalent of an element will, from the definition, be either
the atomic weight itself, or related to it in a simple manner, i.e.,
it will be a simple fraction of it, f , J, f , etc., since 1 atom of the
element combines with 1, 2, 3, etc., atoms of hydrogen, or two
atoms of the element with 3, 5, etc., atoms of hydrogen, and so on.
In only one single case does 1 atom of hydrogen combine with
more than 1 atom of another element, so that in all other cases the
equivalent is either equal to, or less than, the atomic weight. The
exceptional case is hydrazoic acid, a compound of 1 atom of
hydrogen with 3 atoms of nitrogen. The equivalent of nitrogen in
this compound is three times its atomic weight.

Limitations of Dalton's theory. — In its original form, Dalton's
atomic theory provided no means of determining even the relative
weights of the atoms. Thus, although 7 -94 parts of oxygen combine
with 1 part of hydrogen, we do not know how many atoms of each
element the molecule of the resulting water contains. If it con-
tains 1 atom of each element (as Dalton supposed), the atomic weight
of oxygen is 7 -94, but if it contains 2 atoms of hydrogen to 1 atom of
oxygen, as the volume ratio would suggest, the atomic weight of
oxygen is 2 x 7-94 = 15-88.

In general, if Qv Q2 are the weights of two elements which com-
bine together, we must have :

where Al and A9 are the atomic weights, and al} a2 are whole
numbers representing the numbers of atoms of each element,
respectively, which enter into combination. Obviously, a knowledge
°f $1 : $2 alone does not enable us to find the ratio of the atomic
weights, Al : A2) unless the ratio of the numbers of atoms, aa : a2,
is also known.

130 INORGANIC CHEMISTRY CHAP.

Dalton himself was clearly aware of this deficiency, and was
compelled to fall back on empirical rules, which he recognised as
arbitrary. He assumed that, if only one compound of two elements
is known, it is binary, i.e., formed of one atom of each element.
Thus, water was regarded as a compound of one atom each of hydrogen
and oxygen, and ammonia as a compound of one atom each of
nitrogen and hydrogen, since at that time no other compounds of
these elements were known. This rule appears to have been con-
nected with Newton's theory of the repulsion of atoms, according
to which one atom of one element may be attracted by one atom of
another element, but if two atoms of the same element are brought
together they repel each other, and do not form a stable compound
with a third atom. This reasoning had been used before by
W. Higgins ; it is given in Henry's " Chemistry " (1815), and since
Henry was a personal friend of Dalton, he probably derived it from
the latter. Although this rule as to the binary composition of a
single stable compound had, therefore, some theoretical foundation,
Dalton's further rules for the cases where more than one compound
of two elements existed were purely arbitrary.

The work of Berzelius, who extended Dalton's investigations
with great enthusiasm and success, led to no real improvement in
this respect, since his skilful use of chemical analogies, although
leading to the correct results in many cases, was equally arbitrary.
Many chemists, therefore, whilst adopting the experimental basis
of the theory, and the equivalent weights of elements, refused to
attempt to derive the true atomic weights by mere rules. In
particular, Leopold Gmelin, in his large " Handbook " (1817-19 ;
English translation, 19 vols., 1848-72), reverted to the use of
equivalents, and he had numerous followers. In other quarters
an intolerable diversity of systems arose, dictated almost entirely
by authority, and it began to appear as if the atomic theory had
outgrown its usefulness.

Dalton seems to have assumed as self-evident that the particles
of elements in the free state are single atoms. This was the
main source of the difficulties of the earlier theory, since it is
incorrect. The true theory, which would have resolved all
the growing difficulties of Dalton's great generalisation, was
given by the Italian physicist Avogadro, in 1811, but was
entirely unheeded until it was revived in 1858 by his country-
man, Cannizzaro. This theory forms the subject-matter of the
next chapter.

Chemical nomenclature and notation. — The methods of naming
chemical substances constitute chemical nomenclature ; their
representation by symbols is called chemical notation.

The names of the metals are derived from various sources. The
association of the seven metals known to the ancients with the seven

vin THE ATOMIC THEORY 131

planets dates from the Babylonian period, and led to the alchemical
names and symbols :

Gold (yellow) was called Sol (the Sun) and represented by 0 or 0 ;
Silver (white) was named Luna (the Moon), ([ ; Copper was named after
Venus, ? ; Tin was Jupiter 1L ; Iron was named after Mars, the God of
War, $ ; Mercury (mobile) was named after the Messenger of the Gods,
$ ; and Lead (dull and heavy) was Saturn, {7. All these metals,
except mercury, are referred to by Homer ; mercury is first mentioned
by Aristotle (B.C. 384 — 322). The name Mercury still survives, and
silver nitrate is often called lunar daustic.

The nomenclature of the alchemists was purely empirical, and a
name frequently had one meaning to the adept and quite another
to the ordinary man. The same substance had a variety of names,
depending on its mode of preparation. Names were often based
on accidental resemblances. Thus butter of antimony was classed
along with ordinary butter, and oil of vitriol (sulphuric acid) with
olive oil. Such names as liver of sulphur (impure potassium sul-
phide) and cream of tartar (potassium hydrogen tartrate) arose in
this way. Salts were often named after their discoverers, or the
places where they were found (Glauber's salt, Epsom salt).

A scientific nomenclature began with Macquer and Baunie, who,
for instance, classed together the vitriols, or glassy, crystalline
substances : white vitriol (zinc sulphate), green vitriol (ferrous sul-
phate), blue vitriol (copper sulphate). Bergman (1782) invented a
more rational system of nomenclature which indicated the basic
and acidic constituents of salts. E.g., salts of potash, or the
vegetable alkali, were named as follows :

vegetabile vitriolatum (potassium sulphate)
vegetabile nitratum (potassium nitrate).

The modern chemical nomenclature had its origin in a treatise
(" Methode d'une Nomenclature chimique," 1787) drawn up by
Lavoisier, Berthollet, Guyton de Morveau, and Fourcroy, in order
to make the antiphlogistic doctrines less dependent on names
which had arisen during the phlogistic period.

The names of the elements. — Some of the elements (copper, gold,
tin, sulphur) retain their old names ; newly discovered elements
usually have names ending in -um if they are metals, and -on if
they are non-metals (e.g., potassium, and argon). (Selenium and
tellurium were believed to be metals on their discovery ; helium
was named before its isolation.) Many elements have names
derived from Greek roots : chlorine, from xywpos, chloros, greenish-
yellow ; bromine, from /fyw/zos, bromos, a stench ; iodine, from
toeiSrys, io'ides, violet ; chromium, from ^pupa, chroma, colour ;

K 2

132 INORGANIC CHEMISTRY CHAP.

helium, from ^Aio?, helios, the Sun, in the spectrum of which it was
detected. Other elements have been named after mythological
deities or personages : vanadium, from Vanadis, a cognomen of the
Scandinavian goddess Freia ; thorium, from Thor, the Scandinavian
war-god ; tantalum and niobium, from Tantalus and Niobe, of Greek
mythology. The names of places where compounds of elements
were first discovered have sometimes formed the bases of names :
strontium, from Strontian, in Scotland ; ruthenium, from Ruthenia
(Russia) ; ytterbium, from Ytterby (Sweden). Beryllium and
zirconium are named after the minerals, beryl and zircon, which
contain these elements. Palladium and uranium were called after
the stars Pallas and Uranus, discovered about the same time,
whilst selenium and tellurium are named after the Moon (selene)
and the Earth (tellus).

The symbols of the elements. — The present chemical notation is
due to Berzelius (1811), who replaced Dalton's inconvenient circular
symbols by the initial letter, or, in cases where the names of several
elements had the same initial letter, the initial and one other letter,
of the Latin name.

The symbol of an element has a quantitative significance, and
represents one atom, or one atomic weight, of the element. Thus,
H represents 1 part by weight of hydrogen; O represents 15-88
parts by weight of oxygen ; Cl represents 35 -2 parts by weight of
chlorine, and so on. This is the most important feature of the
system of chemical notation.

The symbols of the elements, with their atomic weights, are
given in the table on p. 145. The following are the less obvious

symbols with which the reader should make himself familiar :

ENGLISH LATIN ATOMIC WEIGHT

NAME. NAME. SYMBOL (approximate).

Antimony stibium Sb 120

copper cuprum Cu 63-5

mercury hydrargyrum Hg 200

silver argentum »Ag 107

gold aurum Au 197

iron ferrum Fe 56

lead plumbum Pb 207

potassium kalium K 39

sodium natrium Na 23

tin stannum Sn 119

The symbol W is given to tungsten, from the German name wolfram.
These symbols of the elements are the same in all languages, with the
exception of Az (azote], sometimes used in French for nitrogen. The
names glucinum (Gl) and columbium (Cb) are sometimes used for beryl-

vin THE ATOMIC THEORY 133

Hum and niobium, respectively, from the Greek glukos, sweet-tasting,
and Columbia (America).

The names of compounds.— The names of compounds are formed
from those of their constituents in such a way as to indicate their
composition.

In the names of compounds of two elements, or binary com-
pounds, the name of one element, the more electropositive (p. 275),
comes first, followed by the name of the other element, suitably
contracted and with the termination -ide. The order in which
the elements are taken jn forming the names is as follows :

Metals.

Carbon.

Hydrogen.

Nitrogen, phosphorus, arsenic.

Sulphur, selenium, tellurium.

Halogens (fluorine, chlorine, bromine, iodine).

Oxygen.

E.g., 2 atoms of hydrogen -f- 1 atom of sulphur form hydrogen

sulphide, H2S.

1 atom of sodium + 1 atom of chlorine form sodium chloride,
NaCl.

1 atom of calcium + 2 atoms of carbon form calcium carbide,
CaC2.

2 atoms of sulphur + 2 atoms of chlorine form sulphur chloride,

S2C12.

2 atoms of chlorine + 1 atom of oxygen form chlorine mon-
oxide, C12O.

The formulae of compounds. — These are^ made by writing the
symbols of the elements together, with a small numerical suffix to
indicate how many atoms of each element are present in a molecule
of a compound, as shown in the table above. The number unity is
always understood if no suffix is written.

Since two elements often combine in more than one proportion,
giving different compounds, this is represented in the nomenclature
in one of two ways : (1) by suffixes, (2) by prefixes.

Thus, the two oxides of copper are :

Red oxide of copper, Cu2O, cuprous oxide "I

Black oxide of copper, CuO, cupric oxide J buffixes-

#--f-x —
Two oxides of sulphur are :
SO2, sulphur dioxide "|
S03, sulphur trioxidej Prefixes-

502, sulphur dioxide

503, sulphur trioxide

The suffix -ous denotes the lower, the suffix -ic the higher, pro

134 INORGANIC CHEMISTRY CHAP.

portion of an element. The suffixes are always added to the Latin
names : —

Green chloride of iron, FeCl2, ferrous chloride, or iron dichloride.
Red chloride of iron, FeCl3, ferric chloride, or iron trichloride.

The prefixes sub-, proto-, and sesqui- have practically gone out of use :
Cu2O, copper suboxide, now called cuprous oxide.
CuO, copper protoxide, now called cupric oxide.
FeO, iron protoxide, now called ferrous oxide.
Fe2O3, iron sesquioxide, now called ferric oxide.

In a series of oxides, the one containing the highest proportion of
oxygen is often called a peroxide :

Lead suboxide, Pb2O.

Lead monoxide, Litharge, PbO.

Lead sesquioxide, Pb2O3.

Triplumbic tetroxide, Red lead or minium, Pb3O4.

Lead peroxide, Puce-coloured oxide of lead, PbO2-

It has been proposed to restrict the term peroxide to a special class
of oxides, viz., those giving hydrogen peroxide (H2O2) with acids,
such as Na2O2, BaO2. In this case PbO2 would be called lead dioxide.
The highest oxide of nitrogen definitely known is the pentoxide, N2O5,
which is never called the peroxide, the latter name being very improperly
used for the dioxide, NO2 — apparently because many true peroxides
are dioxides (BaO2, etc.). The name peroxide is, in fact, very loosely
used, and causes great confusion to beginners ; it would seem desirable
to use it only in the restricted sense just explained.

The common names, or special names, are frequently used for
compounds instead of the systematic names. Thus : water (H2O),
ammonia (NH3), hydrazine (N2H4), hydrazoic acid (HN3), sul-
phuretted hydrogen (H2S).

Acids, bases, and salts. — Compounds of three elements are called
ternary compounds ; the most important belong to the classes
known as acids, bases, and salts, containing oxygen. The ter-
minations -ous and -ic are then used to distinguish acids containing
less and more oxygen, the terminations -ite and -ate being used for
the corresponding salts :

Acid. Salt.

Sulphurous, H2SO3 Sodium sulphite, Na2SO3

Sulphuric, H2SO4 Cupric sulphate, CuSO4

Nitrous, HNO2 Potassium nitrite, KNO2

Nitric, HNO3 Lead nitrate, Pb(NO3)2

vm THE ATOMIC THEORY 135

If more than two oxy-acids of an element exist, the prefixes
hypo- (below) and per- (above) are used :

Hyposulphurous acid, H2S2O4 Sodium hyposulphite, Na2S2O4

Sulphurous acid, H2SO3 Potassium sulphite, K2SO3

Sulphuric acid, H2SO4 Lead sulphate, PbSO4

Persulphuric acid, H2S2O8 Potassium persulphate, K2S2O8.

Oxides yielding acids with water are called acidic oxides, or
acid anhydrides (a without ; vSwp (hudor) water) :

SO2, sulphurous anhydride SO3, sulphuric anhydride

P2O3, phosphorous anhydride ; PaO8, phosphoric anhydride

N2O3, nitrous anhydride ; N2O5, nitric anhydride

Oxides yielding bases (alkalies or alkaline earths) with water are
called basic oxides ; they formerly often had names ending in -a,
some of which are still used :

Na2O, soda ; (CaO, lime)

(K2O, potash) ; MgO, magnesia

Li2O, lithia ; BaO, baryta.

By the combination of basic oxides with water, bases are pro-
duced. These contain a metal (or radical, cf. below) united with a
group of atoms OH, called hydroxyl, and they are therefore called
hydroxides (not " hydrates "). Hydroxides of sodium, potassium, and
other so-called alkali-metals are called alkalies ; those of calcium,
strontium, and barium are called alkaline earths.

K20 -f H2O = 2KOH (potassium hydroxide ; caustic potash).
CaO + 2H2O = Ca(OH)2 (calcium hydroxide, slaked lime).

By the combination of acidic oxides with water, acids are pro-
duced : S03 + H20 = H2S04 (sulphuric acid).

Acidic and basic oxides combine to form salts :

SO3 + Na2O = Na2SO4 (sodium sulphate).

Acids and bases also interact to produce salts, but water is at the
same time eliminated :

2NaOH + H2S04 = Na2S04 + 2H20.

The salt Na2S04 may be regarded as sulphuric acid in which two
atoms of hydrogen are replaced by two atoms of sodium. Thus,
acids may be considered as salts of hydrogen, which hydrogen can be
displaced by metals. This takes place directly, for instance, when
metallic zinc dissolves in dilute sulphuric acid :

Zn + H2SO4 = ZnS04 (zinc sulphate) + H2.
Salts are also formed by the action of acids on basic oxides, or

136 INORGANIC CHEMISTRY CHAP.

carbonates ; in the second case gaseous carbon dioxide is evolved
with effervescence :

CuO + H2S04 = CuS04 + H,O
CaC03-|-2HCl = CaCl2 + H3O + C02.

Radicals. — In certain compounds a group of atoms plays the part
of a single atom, and occurs in a whole series of combinations with
other atoms. Thus the salts formed by the combination of ammonia,
NH3, with acids all contain the group NH4, which plays the part
of a metal, and is called ammonium :

NHo + HC1 = NH4C1, ammonium chloride (cf. potassium chloride,

KC1).
2NH3 -+- H2S04 = (NH4)2SO4, ammonium sulphate (cf. potassium

sulphate, K2S04).

Such an unvarying group of atoms present in a series of closely
related compounds is called a radical (Latin, radix, a root). The
group OH (hydroxyl) in bases is a radical.

Chemical calculations. — The systematic notation of chemistry
leads to a great simplification of numerical calculations. The
symbol of an element represents one atom, i.e., a definite weight,
and the formula of a compound represents one molecule, the weight
of which is the sum of the weights of the atoms contained in it.
Calculations of chemically interacting weights are then reduced to
simple proportions.

The notation is also applicable to the representation of the
interaction of elements and compounds ; the resulting expressions
are called chemical equations, and have a quantitative significance.

The formula of a compound is easily found from its percentage
composition, and vice versa. The simplest possible formula derived
from the percentage composition is called the empirical formula.

To find the formula from the percentage composition we
divide the percentage of each element by its atomic weight and
obtain a series of numbers in proportion to the numbers of
atoms of the elements in the molecule of the compound. This
series, reduced to the ratios of the smallest whole numbers, will give
us the empirical formula.

The slight differences from whole numbers often found are due to
experimental errors in the percentage composition. This process of
rounding off must be used with caution : the empirical formula of
cane-sugar is C12H22On, which might be written CH2O, and the
difference put down to experimental errors, if there was not other
evidence that C^H^On is the correct formula.

Since the symbols denote atomic and molecular weights, the
same number of atoms of the same elements must occur (in different

vm THE ATOMIC THEORY 137

groupings, it is true) on each side of a chemical equation ; or, as
is said, the equation must balance.

SUMMARY OF CHAPTER VIII

The laws of stoichiometry are explained by a hypothesis called the
atomic theory. This supposes that: (1) all matter is made up of
minute particles, called atoms, which are undivided in chemical changes ;
(2) the atoms of each element are identical ; (3) in chemical combination
a whole number of atoms of one element is associated with a whole
number of atoms of another element, or elements, to form a molecule
of the compound.

Each atom has a definite, but exceedingly small, weight or mass.
The absolute mass of the lightest atom, viz., that of hydrogen, is
1*66 X 10- 24 gm. The ratio of the weight of an atom of any
element to the weight of the hydrogen atom is called the atomic weight
of the element. Each element has a symbol, denoting one atomic
weight of the element. The formula of a compound, denoting one
molecular weight, contains the symbols of its constituent elements
written side by side, each with a numerical suffix indicating how many
atoms of that element occur in the molecule of the compound.

EXERCISES ON CHAPTER VIII

1. Explain what is meant by a hypothesis. Give a short account
of the origin and content of the atomic hypothesis iised in chemistry.
Mention any other hypotheses with which you are acquainted.

2. From what sources are the names of the chemical elements
derived ? How are they denoted by symbols ?

3. Describe carefully, with the use of an example, what are the
separate steps involved in a chemical calculation. On what experi-
mental and theoretical results are these steps based ?

4. Barium peroxide has the formula BaO2. On heating, it evolves
oxygen gas, with the formula O2, and leaves a residue of baryta, BaO.
Write the chemical equation of the reaction, and, from the result that
31-76 gm. of oxygen at S.T. P. occupy 22-3 litres, find the weight of
barium peroxide required to make 10 litres of oxygen, measured at
15° and 740 mm.

5. Potassium dichromate has the formula K2Cr2O7. On heating
with concentrated sulphuric acid it gives off oxygen gas and water
vapour, and leaves a residue containing potassium sulphate, K2SO4,
and chromium sulphate, Cr2(SO4)3. Write down the equation of the
reaction, and find how many litres of oxygen, measured at 10° and
762 mm., are evolved from 1 kgm. of dichromate.

6. What weights of crystallised potassium ferrocyanide and of
concentrated sulphuric acid, sp. gr. 1-84, are required to prepare 100 gr.
of carbon monoxide (p. 702) ? Sulphuric acid of sp. gr. 1-84 contains
96-6 per cent. H2SO4.

CHAPTER IX

AVOGADRO'S HYPOTHESIS AND THE MOLECULE

The law of gaseous volumes.— Reference has been made to the
determination of the relative combining volumes of hydrogen and
oxygen, which were found by Cavendish to be very nearly 2:1.
Alexander von Humboldt and Joseph Louis Gay-Lussac in 1805
confirmed this result, and the latter, impressed by the simplicity
of the ratio, extended the researches to other chemical reactions
between gases. In 1808 he published the results, and from them
deduced the following law :. When chemical changes occur between
gases, there is always a simple relation between the volumes of the inter-
acting gases, and also of the products, if these are gaseous. The same
conditions of temperature and pressure are assumed.

EXAMPLES.

1 volume of oxygen combines with 2 volumes of hydrogen to give
2 volumes of steam.

1 volume of chlorine combines with 1 volume of hydrogen to give
2 volumes of hydrochloric acid.

2 volumes of carbonic oxide combine with 1 volume of oxygen to
give 2 volumes of carbonic acid.

2 volumes of nitrogen combine with 1 volume of oxygen to give
2 volumes of nitrous oxide.

1 volume of nitrogen combines with 1 volume of oxygen to give
2 volumes of nitric oxide.

1 volume of nitrogen combines with 2 volumes of oxygen to give
2 volumes of nitrogen dioxide.

1 volume of nitrogen combines with 3 volumes of hydrogen to give
2 volumes of ammonia.

Later experiments (p. 213) show that the law is not quite
exact. Burt and Edgar found the combining volumes of hydrogen
and oxygen to be 2-00288 : 1 ; Gray and Burt from 2 volumes of
hydrochloric acid gas obtained 1 -0079 volumes of hydrogen ;

138

CH. ix AVOGADRO'S HYPOTHESIS AND THE MOLECULE 139

Guye and Pintza showed that 1 volume of nitrogen combines with
3-00172 volumes of hydrogen to form ammonia. All these numbers
refer to S.T.P. The slight differences from whole numbers appear
to be due to the different compressibilities of the gases, i.e., the
deviations of the gases from Boyle's law.

Thus, if 100 c.c. of oxygen at S.T.P. are converted into carbon
dioxide by burning carbon in the gas, the volume is found to have
contracted slightly. Carbon dioxide is more compressible than oxygen,
and occupies a slightly smaller volume than that of the oxygen it
contains.

Gay-Lussac remarked that, since gases combine by weight in
atomic proportions, or simple multiples of these, and by volume in
simple ratios, there must be some simple relation between the
atomic weights and the combining volumes. Berzelius made the
assumption that equal volumes of elementary gases contain equal
numbers of atoms. Dalton objected strongly to this statement.
In the first place, his own (inexact) measurements of combining
volumes did not confirm Gay-Lussac's law : thus, he found that 1 -97
volumes of hydrogen combine with 1 volume of oxygen. In the
second place, he pointed out that the density of a gas is not the
same thing as the weight of its ultimate particle ; steam, for
instance, is lighter than oxgyen, whereas the ultimate particle of
steam must be heavier than that of oxygen, since it contains the
latter.

A more serious difficulty was also pointed out by Dalton. One
volume of oxygen combines with 1 volume of nitrogen to produce
2 volumes of nitric oxide. Now if 1 volume of oxygen (say 1 litre)
contains n atoms, then 1 litre of nitrogen will also contain n atoms.
Combination occurs between equal volumes, therefore, according
to the above theory, atom for atom ; hence there will be n mole-
cules of nitric oxide produced. But these are found to occupy a
volume of 2 litres, hence nitric oxide can contain only half as many
particles in a given volume as nitrogen or oxygen. Avogadro in
1811 set out to explain this discrepancy, and he succeeded in
clearing away all the difficulties by a simple assumption which,
when it was made, appeared almost obvious.

Avogadro's Hypothesis. — Avogadro began by assuming that the
simple hypothesis of equal numbers of particles in equal volumes
is correct. The discrepancies must then arise from an incorrect
method of applying the hypothesis to the experimental results.

Avogadro's hypothesis, that equal volumes of all gases and vapours, under
the same conditions of temperature and pressure, contain identical numbers of
molecules, shows that " the ratios of the masses of the molecules are

140 INORGANIC CHEMISTRY CHAP.

the same as those of the densities of the different gases at equal
temperature and pressure."

By a molecule is meant the smallest mass of a substance capable of
existing in the free state.

In the case of gases we shall see (Chapter XV) that the molecules
are in motion, and Maxwell has defined a molecule as that small portion
of matter which moves about as a ivhole so that its parts, if it has any,
do not part company during the motion of agitation of the gas. There
is reason to believe that the constitution of liquids is similar to that
of gases, but the molecules are closer together and glide over one
another. In solids, the molecules probably perform small oscillations
about stationary positions (p. 271).

Molecules of gases. — The difficulty which had confronted Gay-
Lussac and Berzelius was now cleared away. Avogadro pointed
out that the molecules of elementary gases are not necessarily the
atoms themselves, but usually consist of groups or clusters of atom,s,
moving about as though they were single particles. Both kinds of
particles, atoms and molecules, had been called '; atoms " by Dalton,
but they were really different. Avogadro arrived at this important
conclusion as follows.

Chlorine and hydrogen combine in equal volumes to form a volume
of hydrochloric acid equal to the sum of the volumes of the elemen-
tary gases. Equal volumes of chlorine and hydrogen, however,
contain identical numbers of molecules, say n. The 2 volumes of
uncombined mixed gases will therefore contain 2 n molecules, of
which n are of hydrogen, and n are of chlorine. After combination,
the 2 volumes of hydrochloric acid must by hypothesis also contain
2 n molecules. Now each molecule of hydrochloric acid must contain
at least one atom each of chlorine and hydrogen, hence there must
be at least 2 n atoms of each element present. Thus, the n mole-
cules of chlorine gas, and the n molecules of hydrogen gas, must
each have contained 2 n atoms ; in other words, a molecule of each of
the elementary gases must contain at least two atoms.

Volume diagrams. — This reasoning is most clearly grasped by the
use of volume diagrams, in which the volumes of the gases are
represented by squares or rectangles, and the molecules by small
circles. It must be emphasised that Avogadro's hypothesis does
not assert that the volumes of the actual molecules themselves are
equal, but only that the volumes of the gases which contain equal
numbers of molecules are identical. The compressibility of gases,
and the relatively small volumes to which they are reduced by
liquefaction, show that there are large spaces between the mole-
cules in a gas, and the different volumes of liquid obtained from

AVOGADRO'S HYPOTHESIS AND THE MOLECULE

141

equal volumes of different gases indicate that the actual molecules
of different gases have different sizes. All the gases will be con-
sidered at the same temperature and pressure.

EXAMPLE 1. — Combination of hydrogen and oxygen (Fig. 73). —
2 volumes of hydrogen + 1 volume of oxygen = 2 volumes of steam.

/. 2 n molecules of hydrogen + n molecules of oxygen = 2 n mole-
cules of steam.

4 n atoms of hydrogen -f 2 n atoms of oxygen = 2 n molecules of
steam (containing 6 n atoms).

•<?

•s

«•

//2 //2 02 //20 W20

FIG. 73. — Diagram illustrating combination of Hydrogen and Oxygen.

EXAMPLE 2. — Combination of hydrogen and chlorine (Fig. 74). —
I volume of hydrogen + 1 volume of chlorine = 2 volumes of hydro-
chloric acid.

/. n molecules of hydrogen + n molecules of chlorine = 2 n mole-
cules of hydrochloric acid.

2 n atoms of hydrogen + 2 n atoms of chlorine = 2 n molecules of
hydrochloric acid (4 n atoms).

•0 •©

•o

•0

H2 C12 HGI HCI

FIG. 74. — Diagram illustrating combination of Hydrogen and Chlorine.

EXAMPLE 3. — The combustion of carbon in oxygen to form carbon
Lioxide (Fig. 75). — In this case we know nothing of the composition of
the molecules of carbon, since these are present in a solid, to which
Avogadro's hypothesis does not apply. If we assume that one atom
>f carbon combines with two atoms of oxygen to form a molecule

carbon dioxide, the fact that no change in volume occurs when

rbon burns in oxygen is explained, but it is clear that the same
jsult is obtained if we assume that n atoms of carbon combine with

atoms of oxygen to produce a molecule of carbon dioxide. The

142

INORGANIC CHEMISTRY

CHAP.

only result which may be deduced directly is that a molecule of carbon
dioxide contains a molecule of oxygen (2 atoms).

Solid _
Carbon

C02

FIG. 75. — Diagram illustrating combination of Carbon and Oxygen.

Molecular weight and relative density. — The molecular weights of
substances which exist in the gaseous state, or can be* converted
into vapours, may be compared by finding the ratio of the densities.
The weight of any volume of the gas or vapour is compared with
the weight of an equal volume of a standard gas under the same
conditions :

Mol. wt. of substance __ Wt. of any vol. of substance
Mol. wt. of standard ~~Wt. of equal vol. of standard'
by Avogadro's hypothesis.

The molecular weight of a substance is defined as the sum of the
atomic weights of its constituents :

Wt. of a molecule of substance
Molecular weight = — z — - , , , — — .

Wt. ot an atom ot hydrogen

The relative density of a gas or vapour has been defined as :

Wt. of anv vol. of gas or vapour

Relative density = -T7, =-^ , . & , , — ,

Wt. ot equal vol. ot hydrogen

both substances being under the same conditions. But equal
volumes contain equal numbers of molecules :

Wt. of n molecules of substance

Relative density =

Wt. of n molecules of hydrogen
Wt. of one molecule of substance

Wt. of one molecule of hydrogen

.*. Molecular weight = relative density X molecular weight of
hydrogen.

The molecular weight of hydrogen is defined in the same way as
other molecular weights, viz., as the ratio of the weight of a molecule
of hydrogen to that of an atom of hydrogen. Since gaseous com-
pounds of hydrogen, when formed from hydrogen gas, may occupy
double the volume of the latter, but never occupy more than double
the volume, it may be assumed that the molecule of hydrogen

ix AVOGADRO'S HYPOTHESIS AND THE MOLECULE 143

consists of two atoms ; in other words, the molecular weight of
hydrogen is two. Thus :

Molecular weight = Relative density x 2.

Determination of atomic weight from gas or vapour densities.—

Cannizzaro in 1858 showed that Avogadro's hypothesis can be
systematically applied in the determination of atomic weights.
By means of vapour density measurements, the molecular weights
of as many volatile compounds of an element as possible are found.
By analysis, it is then found what weights of the particular element
are contained in the molecular weights of the various com-
pounds. These must be whole multiples of the atomic weight,
and, if the number of compounds taken is large enough, at least one
of the weights of the element present in the molecular weights of its
compounds will probably be the atomic weight itself.

The atomic weight of an element is the least weight of the element
contained in a molecular weight of any of its compounds.

It will be seen that this is not an independent definition of atomic
weight, but is merely a consequence of the molecular theory.

It cannot be too strongly emphasised that the determination of
the relative density of one compound of an element, or of the
element itself if it is volatile, can give no sure indication of the
atomic weight. The molecules of the particular compound selected,
and those of the vapour of the free element, may contain two, three,
or any number of atoms of the element, for all we know to the
contrary. The larger the number of compounds investigated, the
greater is the probability that at least one contains only one atom
of the element in a molecule.

The method used by Cannizzaro may be illustrated by a table of
oxygen compounds. The numbers are approximate only.

OXYGEN COMPOUNDS.

Compound. Rel. density Mol. wt. Wt. of oxygen in one

(H = 1)A = 2 x A mol. wt. of compound.

Oxygen gas 16 32 16

Water 9 18 16

Carbon monoxide 14 28 16

Carbon dioxide 22 44 16 X 2

Sulphur dioxide 32 64 16 X 2

Sulphur trioxide 40 80 16x3

Alcohol 23 46 16

Ether 37 74 16

Nitrous oxide 22 44 16

Nitric oxide 15 30 16

144 INORGANIC CHEMISTRY CHAP.

The least weight of oxygen found in a molecular weight of any
one of these compounds is 16, and hence this must be taken as
the atomic weight.

A molecule of water contains one atom of oxygen, of weight 16,
and therefore contains 18 — 16 = 2 parts, or two atoms, of hydrogen.
The formula of water is therefore H20. In this way the problem
which had eluded Dalton, of finding the number of atoms of the
elements in the molecule of a compound, is easily solved.

Similarly, a table of carbon compounds may be drawn up.

CARBON COMPOUNDS.

Compound. Bel. density Mol. wt. Wt. of carbon in one

(H = 1) A = 2 x A mol. wt. of compound.

Methane 8 16 12

Ethane 15 30 12 x 2

Ethylene 14 28 12 x 2

Alcohol 23 46 12 x 2

Ether 37 74 12 x 4

Benzene 39 78 12 X 6

Carbon monoxide 14 28 12

Carbon dioxide 22 44 12

The atomic weight of carbon deduced from these results is 12.
Thus, in 78 parts of benzene there are 72 parts, or 6 atoms, of
carbon. Hence there are 78 — 72 = 6 parts, or 6 atoms, of
hydrogen. The formula of benzene is thus C6H6.

The molecular weights found from the relative densities are only
approximate, since the vapours do not accurately obey the gas laws,
and the determinations are usually made only roughly. The accurate
values of the atomic and molecular weights are found from the
refined chemical analyses of the compounds, and the vapour
density measurements used simply to decide between various possible
molecular weights (see p. 147).

By drawing up tables similar to the above for as many elements
as possible, we arrive at the atomic weights of these elements. In
some cases, an element does not form volatile compounds, so that
the method cannot be applied. Alternative methods must then be
used, which are described in the next section.

Confirmation of atomic weights. — The atomic weights derived
from the relative densities have been confirmed by a variety of
independent methods. These remove the possibility that the
least weight of an element found in the molecular weights of all the
compounds examined may still be a multiple of the atomic weight,
since it is very improbable that all the independent methods

AVOGADRO'S HYPOTHESIS AND THE MOLECULE

145

INTERNATIONAL ATOMIC WEIGHTS (1921).

Atomic weight.

Element. Symbol
Aluminium ... Al

Antimony

Argon

Arsenic

Barium

Beryllium

Bismuth

Boron B

Bromine .
Cadmium
Cyesium .
Calcium .
Carbon
Cerium
Chlorine .
Chromium

Cobalt Co

Copper Cu

Dysprosium ... Dj
Erbium... . Er

Europium

Fluorine

Gadolinium

Gallium

Germanium

Gold .. . Au

Helium
Holmium..
Hydrogen.
Indium

Iodine

Iridium Ir

Iron Fe

Krypton KT

Lanthanum ... La
Lead* .. . Pb

Lithium .....
Lutecium .....
Magnesium ..
Manganese . .
Mercury .....
Molybdenum

.H - 1
26-8

119-2
39-6
74-37

136-28
9-0

206-4
10-8
79-29

111-51

131-76
39-75
11-910

139-15
35-18
51-6
58-50
63-07

161-2

166-4

150-8
18-9

156-1
69-5
71-9

195-6
3-97

162-2
1-000

113-9

125-91

191-6
55-40
82-26

137-9

205-55
6-89

173-6

24-13

54-49

199-0

95-2

O = 16
27-1

120-2
39-9
74-96

137;37

208-0
10-9
79-92

112-40

132-81
40-07
12-005

140-25
35-46
52-0
58-97
63-57

162-5

167-7

152-0*-
19-0

157-3
70-1
72-5

197-2
4-00

163-5
1-008

114-8

126-92

193-1

55-84

82-92

139-0
207-20
6-94

175-0

24-32

54-93
200-6

96-0

Atomic weight.

Element. Symbol. H = 1
Neodymium... Nd 143-2

Neon Ne 20-0

Nickel Ni 58-21

Niobium Nb 92-4

Niton Nt 220-6

Nitrogen N 13-897

Osmium

Oxygen

Palladium . . .
Phosphorus

Platinum

Potassium . . .
Praseodymium Pr
Radium . . .
Rhodium ...
Rubidium
Ruthenium
Samarium
Scandium

Selenium

Sc
Se

Silicon Si

Silver Ag

Sodium
Strontium
Sulphur
Tantalum
Tellurium
Terbium ,
Thallium ,
Thorium .
Thulium .

Tin

Titanium .
Tungsten .
Uranium .
Vanadium

Xenon Xe

Ytterbium
Yttrium .

189-4
15-87
105-9
30-79
193-6

38-79
139-8
Ra 224-2
Rh 102-1
Rb 84-77
Ru 100-9
Sa 149-2
44-7
78-6
28-1
107-04
22-82
86-93
31-81
180-1
126-5
Tb 157-9
Tl 202-4
Th 230-31
Tm 167-2
Sn 117-8
47-72
182-5
236-3
50-6
129-2
Yb 172-1
Yt 88-62

Zinc

Zirconium

Zn
Zr

64-85
89-9

O = 16
144-3

20-2

58-68

93-1
222-4

14-008
190-9

16-00
106-7

31-04
195-2

39-10
140-9
226-0
102-9

85-45
101-7
150-4

45-1

79-2

28-3
107-88

23-00

87-63

32-06
181-5
127-5
159-2
204-0
232-15
168-5
118-7

48-1
184-0
238-2
51-0
130-2
173-5
89-33
65-37
90-6

146 INORGANIC CHEMISTRY CHAP.

should agree with this particular multiple. These methods will be
considered in more detail later ; a brief summary only is given here.

1. The ratio of the specific heats of a gas or vapour at constant pres-
sure, cp, and at constant volume, cv, respectively, viz., cp/cv, has, ac-
cording to the kinetic theory (p. 598), the value 1-667 only when the
molecule consists of a single atom. In 1875 Kundt and Warburg
found that cp/cv had the value 1-667 for mercury vapour, hence the
molecules of the latter consist of single atoms. The relative density
of mercury vapour is 100, hence the molecular weight is 200. This,
however, must in the present case be equal to the atomic weight. It
was found, in fact, that 200 parts of mercury was the least weight ever
contained in a molecular weight of the volatile compounds. If the atomic
weight found by the latter method can be shown in one case, viz., mercury*
to be the real atomic weight, and not a multiple, one may reasonably
assume that in other cases also the method gives the real atomic weights.

2. Dulong and Petit in 1819 found that the product of the atomic
weight and the specific heat of a solid element is approximately
constant, and equal to 6-3. Hence if the specific heat of a solid element
is determined, and 6-3 is divided by this number, we obtain an approxi-
mate value of the atomic weight. This may be used to check the value
found by the vapour density method.

3. Mitscherlich in 1819 found that compounds having analogous
formulae crystallise in the same form, or are isomorphous. Tims,
potassium chromate crystallises in the same form as potassium
sulphate. The atomic weight of * sulphur is found from the
compositions and densities of its volatile compounds to be 32. The
atomic weight of potassium is found from the specific heat to be 39.
Thus the formula of potassium sulphate is found to be K2SO4. From
its isomorphism with the sulphate we assume that the formula of the
chromate is K2CrO4, and hence, from an analysis of the compound,
we find the atomic weight of chromium to be 52. This is confirmed
by the specific heat of the metal.

4. The formulae of compounds which show undoubted similarities
in chemical properties are usually similar. Thus, the oxides of iron,
aluminium, and chromium are given the similar formulae, Fe2O3,
A12O3, and O2O3. If the atomic weight of chromium is found, as
above, those of aluminium and iron can readily be determined. This
method is the least trustworthy of all. Thus, beryllium oxide is
similar in practically all its chemical properties to aluminium 'oxide,
but has the formula BeO.

5. The position of the element in the Periodic system (Chap. XXIV)
is probably the most convincing proof that the present values of the
atomic weights are the correct multiples. No other values would place
the elements in their correct positions.

ix AVOGADRO'S HYPOTHESIS AND THE MOLECULE 147

Molecular weights of elements. — It has been emphasised that the
relative density of an element itself gives no indication of the value
of the atomic weight. Molecules of elements in the gaseous state
may contain from one to eight atoms :

Monatomic : Hg, Na, K, Zn, Cd, He, A, Ne, Kr, Xe, Nt, I, Cl(?), Bi.

Diatomic : H2, O2, N2, C12, Br2, I2, F2, S2, Se2, Te2, As2, Sb2(?), Bi2.

Triatomic : O3.

Tetratomic : P4, As4.

Hexatomic : S6 (?).

Octatomic : S8.

The absence of the types X6 and X7 is noteworthy.

Limiting densities. — A comparison of the normal densities of gases
cannot give exact ratios of the molecular weights, even when the
most accurate values of the densities are used. For, even if equal
volumes of different gases contained accurately equal numbers of
molecules at one particular pressure, these volumes would, on
account of the slightly different compressibilities of the different
gases, not remain exactly equal at another pressure. The numbers
of molecules in these unequal volumes would, however, still be
equal.

The unequal compressibilities of gases result from the deviations
from Boyle's law, since the latter gives equal compressibilities for
all gases. Since the deviation from Boyle's law becomes less and less
as the pressure becomes smaller, and appears to vanish at very small
pressures, it may be assumed that the ratio of the densities at very
low pressure, or the ratio of the limiting densities (^>-»0), wall give
the exact ratio of the molecular weights (D. Berthelot, 1899).

If a mass W gm. of gas occupies at 0° a volume v litres ufider a
pressure p atm., we may call the quotient W/pv the density per
unit pressure. If the gas obeyed Boyle's law, this would be the
same at all pressures, since then pv = const. Owing to deviations
from Boyle's law, the quotient depends on the pressure. If p = 1,
we have the normal density ; if p -> 0 the quotient approaches the
value for an ideal gas, which is called the limiting density. The
ratio of the limiting densities of two gases is the ratio of the mole-
cular weights :

M - M Wa . Wb

Ma - JyJ-b = - - T-

where p0v0 is the limiting value of pv as^) -> 0.

If D is the normal density of a gas, D = - , where p^ is the

Pi^i
value of pv for p = 1, .

/. limiting density = normal density X (•£&].

\1VV

L 2

148 INORGANIC CHEMISTRY^ CHAP.

The ratio P^/PQVQ may be determined for any arbitrary mass of
gas by two methods :

(i) For gases which deviate only slightly from Boyle's law between
zero pressure and 1 atrn., (pQvQ — Jw)/Po*i>> °r the relative deviation
from Boyle's law, may be assumed to be proportional to the pressure :
(Povo — pv}IP X Povo — const. = A. This is called the compressi-
bility coefficient. Its value may be found from two measurements
of pv between 1 atm. and zero pressure. Thus, p^ = pQvQ( 1 — A ),

• normal density
since «i = 1, ^.e., limiting density = — — -. -- *..

1 — A.

(ii) From a number of measurements of pv, a curve can be drawn
in which pv is plotted against p. Extrapolation to p — 0 gives
the value of £>0v0, and then the limiting density is found by multiplying

the normal density by

EXAMPLE 1.— The atomic weight of oxygen from the relative density.

Normal density. Compressibility = J. = l — p^/Wo

Hydrogen 0-089873 + 0-00054

Oxygen 1-42906 -0-000964

Limiting density of hydrogen =0-0898 7 3 X-, — ^ * . . =0 -089922 gm./lit.

— '

Limiting density of oxygen = 1 -42906 x fXoo096l = J '42768 g^/h't.

The ratio of the limiting densities is equal to the ratio of the mole-
cular, or in this case the atomic, weights ; hence :

Atomic weight of oxygen = 1-42768/0-089922 = 15-877.

The number found by direct synthesis of water (p. 64) is 15-879.

EXAMPLE 2. — The atomic weight of chlorine from the density of
hydrochloric acid (Gray and Burt).

Normal density. p^ pQvQ

Hydrogen chloride 1-63915 58403 55213 (extrapolated).

Limiting density of HC1 = 1-63915 X 54803/55213 = 1-62698,
Molecular weight of HC1 =2 X 1-62698/0-089922 =36-187,
Atomic weight of Cl = 36-187 — 1 = 35-187.

The value found by Edgar by direct synthesis of HC1 is 35-186,
By heating aluminium in 2 volumes of hydrogen chloride, measured

at S.T.P., 1-00790 volumes *f hydrogen were obtained. The molecular

weight of HC1 is therefore :

1-63015 ' 2

0-089873 1-00790

agreeing to about 1 part in 10,000 with the value from the limiting
density.

ix AVOGADRO'-S HYPOTHESIS AND THE MOLECULE 149

These examples show that the method of limiting densities gives
results at least as accurate as those found by chemical methods.
In some cases, greater accuracy is probably attained by the density
method.

Gram-molecular volume. — The molecular weight in grams of any
substance is called the gram-molecular weight, or sometimes the mol.
In the case of gases, Avogadro's hypothesis shows that, at a given
temperature and pressure, the gram-molecular weight will occupy
a constant volume. At S.T.P. (0° and 760 mm.) this is called the
gram-molecular volume, or sometimes the molar volume.

This value is the same for all gases only if the latter are in the
ideal state : it may be calculated with close approximation from the
normal density of hydrogen :

Gram-molecular volume = volume of 2 grams of hydrogen at S.T.P.

2 2

_ 99 .9

normal density 0-089873
The accurate value, for an ideal gas, is obtained from the limiting
density of hydrogen :

2
Gram-molecular volume = Q .989922 = 2^*242 litres.

In this book the value 22-24 litres (H = 1) will be adopted.

The gas constant. — The general equation for an ideal gas is :
pv/T = constant. For a gram-molecular weight of an ideal gas at
S.T.P.: p = l atm., v = 22 -24 litres, T = 273-09 .'. the value
of the constant in the above equation is 22-24/273-09 = 0-08145.
This number, which is the same for a gram-molecular weight of any
gas in the ideal state, is called the gas constant, and is denoted by
R. Thus, the general gas equation, for a gm. mol. of an ideal gas, is
pv/T = R, or pv = RT, where R = 0-08145 if p is in atm., v in
litres, and T is the absolute temperature Centigrade.

•The value of R in absolute units may be calculated as follows :
p = 1 atm.= 1-01313 X 107 dynes per sq-. cm. ; v = 22-242 litres =
22242 c.c. ; T = 273-09°. /. R = 8'252 x 1(F ergs per 1°. In heat
units, the value of R is obtained by dividing the value in ergs per
degree by the mechanical equivalent of heat. J = 4 '186 X 10r ergs
per gram calorie, .'. R/J= 1*971 gm. cal. per 1°.

The volume occupied by n gm. mol. of a gas is n times that
occupied by one gm. mol. under the same conditions. Thus, if
V is the volume of n gin. mol. of gas, the general equation becomes
pV = n RT. If the weight of the gas is W gm., and the molecular
weight is M , n = W/M.

In calculations involving gaseous volumes, one may use either
the general gas equation, pv = nRT, or, more conveniently, the
relation that 1 gm. mol. at S.T.P. occupies 22-24 litres. The

150 INORGANIC CHEMISTRY CHAP.

equations must be written so as to express reactions between
molecules of the substances, since only in this case are the volume
relations correctly given.

Thus, the equation H -f- Cl = HC1, although it gives the correct
weight ratios, does not give the correct volume ratio. This is
expressed by the molecular equation : H2 -f C12 = 2HC1.

If solids or liquids participate in the reaction, their volumes are
neglected, since Avogadro's law does not apply to them.

EXAMPLE 1. — Find the volume of 100 gm. of chlorine at 15° and
5-4 atm. pressure.

(a) From the gas equation : pv = nRT.

p = 5-4 atm. ; T = 273 + 15 = 288 ; R = 0-08145 lit. atm./degree ;
n = 100/70-4 ;

0-08145 X 100 X 288
thus v = 5-4 X 70-4~ = 6'17 litres-

(b) F-rom the molecular volume :

1 gm. mol., or 70-4 gm., of chlorine occupies 22-24 litres at S.T.P.
100 gm. of chlorine = 100/70-4 gm. mol., and at 15° and 5-4 atm.
this will occupy :

100 x 22-24 X 288 X 1

70-4 x 273 X 5-4 = 6-17 litres.

In both cases the number is approximate, since chlorine departs
considerably from the ideal state under the given conditions.

Abnormal vapour densities. — Acetic acid has the empirical formula
CH2O, and its vapour density at 250° under 760 mm. pressure is
29 (H = 1), hence the molecular weight is of the order of 58. But
02H4O2 = 60, hence under these conditions the vapour has this for-
mula. At lower temperatures, under 760 mm. pressure, the density
is greater — e.g., at 125° it is 44-5, corresponding with a molecular
weight of 89, which approximates to C3H6O3 = 90. The density
also increases with the pressure when the temperature is constant,
and the change with temperature or pressure occurs gradually, so
that it is only at isolated points that the density corresponds with
a chemical formula.

Playfair and Wanklyn (1862) pointed out that this apparent
exception to Avogadro's law could be explained on the assumption
that the vapour of acetic acid below 250° was a mixture of the normal
molecules, C2H4O2, with molecules of greater density, C3H6O3 or
C4H8O4. The substance is then said to be associated. By rise of
temperature, or decrease of pressure, the associated molecules
gradually break up into the normal molecules : (C2H402)2 z±: 2C2H4O2.
It is probable that these associated molecules exist in the liquid
acid.

A different behaviour is shown by another group of substances, of

ix AVOGADRCVS HYPOTHESIS AND THE MOLECULE 151

which ammonium chloride is typical. This salt is produced by the
direct union of ammonia, NH3, with hydrochloric acid, HC1, and its
simplest formula is thus NH3,HC1, or NH4C1 = 53-2. Bineau,
however, found its vapour density to be only 124, giving a
molecular weight of 24-8. This is roughly half the least possible
theoretical value, and corresponds with the formula NJH2C1J.
This and similar deviations (phosphorus pentachloride, ammonium
carbamate, etc.) led Deville to question the validity of Avogadro's
law, but the true explanation was put forward simultaneously and
independently by Kopp, Kekule, and Cannizzaro in 1857-8.

Dissociation by heat. — Mitscherlich in 1833 had observed that
antimony pentachloride on volatilisation by heat breaks up partially
into antimony trichloride and free chlorine : SbCl5 = SbCl3 -f- C12.
The two constituents recombine on cooling, but can be separated
from the mixture by their different volatilities. Since the reaction
is reversible, it may be written : SbCl5 ^± SbCl3 + C12. Reactions
of this type are examples of thermal dissociation, i.e., the gradual
decomposition of a compound by heat, in such a way that the pro-
ducts of decomposition recombine on cooling. They differ from
such reactions as the decomposition of potassium chlorate by heat,
as the products of these remain uncombined even after cooling.

J. H. Gladstone (1849) also found that the pale yellow solid
phosphorus pentabromide partially dissociates when heated into the
vapour of the tribromide (colourless) and free bromine (red) :
PBr5 ^ PBr3 -f Br2. The vapour is red, owing to the presence of
free bromine. If the vapour is contained in an open flask, bromine
diffuses out, and the denser PBr3 remains. It was therefore reason-
able to assume that ammonium chloride also, on heating, breaks up
into ammonia and hydrogen chloride : NH4C1 z=± NH3 -f- HC1,
which recombine on cooling. The density would then, for
complete decomposition, be half the theoretical density, because the
decomposed gas occupies double the volume it would if no decom-
position had taken place.

Pebal (1862) was able to confirm this assumption by separating
the two gases, NH3 and HC1, from the vapour by diffusion. Am-
monia is much lighter than hydrochloric acid and therefore diffuses
more rapidly (cj. p. 191).

Pebal used the apparatus shown in Fig. 76. The tube D con-
tained a plug of asbestos, c, and above this was placed a piece of
sal ammoniac (NH4C1), d. The tube was enclosed in a wide test-
tube, contained in a jacket heated in a charcoal furnace. Hydrogen
was passed in through the tubes a, 6, on both sides of the plug, and
escaped through tubes to A and B, containing pieces of blue and red
litmus paper, respectively. The red litmus was turned blue,
because ammonia escaped more rapidly through the asbestos plug
than the hydrochloric acid ; the latter was swept out through the

152 INORGANIC CHEMISTRY CHAP.

other tube, and turned the litmus red. Deville objected to this
experiment, on the ground that the vapour might have been decom-
posed by contact with the asbestos plug ; Than (1864) then replaced

the latter by a plug of
solid sal ammoniac (Fig.
77) and obtained the
same result, so that the
dissociation of ammonium
chloride was proved.

Marignac (1868) then
showed that the absorp-
tion of heat required
to volatilise ammonium
chloride is practically
equal to the heat evolved
when the gases ammonia
and hydrochloric acid
combine together to pro-
duce the former com-
pound, and hence the
compound must split up
into the two gases on
volatilisation.

In dissociation, as in
association, the change
occurs gradually, so that
a state of chemical equi-
librium is established, in which the dissociating substance and the
products of dissociation exist side by side : NH4C1 zn NH3 + HC1.
The extent of dissociation increases as the temperature rises, as is
seen, for instance, in the pro-
gressive darkening in colour of
the vapour of phosphorus
pentabromide.

The dissociation of the
colourless hydrogen iodide gas
on heating may be seen from
the violet colour of the iodine
vapour produced :
2HI =± H2
Determination of the extent FIG 77._Tnan'S Experiment on the DISSO-

Of diSSOCiation from the Vapour elation of Ammonium Chloride.

density. — The degree of dissocia-
tion, y, i.e., the fraction of the total number of molecules dissociated
under given conditions, can in many cases be deduced from the
vapour density of the substance. This method, however, is not

FIG. 76. — Pebal's Experiment on the Dissociation of
Ammonium Chloride.

1-7
_ i_ i

ix AVOGADRO'S HYPOTHESIS AND THE MOLECULE 153

applicable to cases where there is no change of volume on dissocia-
tion, e.g., HI -f- HI ^ H2 -}- I2. In such cases the degree of
dissociation must be determined by other methods (p. 348).

In the dissociation of substances such as phosphorus pentachloride,
when a change of volume occurs :

PC15 ^± PC13 -f C12
1 vol. 2 vols.

the progress of the dissociation may be followed by the vapour
density.

Let each molecule of the initial substance break up into x mole-
cules on dissociation. Then if
y is the degree of dissociation
under given conditions, we shall
have in the gas, if N molecules
of substance are taken :

N(l — y) molecules of original
substance,

Nxy molecules of the products FIGK 73.— Diagram illustrating Dissociation.
of dissociation.

The number of molecules before dissociation is N ; that after
dissociation is N (1 - y) + Nxy = N[l -f y(x — 1)] (see Fig. 78).
By Avogadro's law :

Volume after dissociation N [1 + y(x — 1)]

Volume before dissociation N

The densities are inversely proportional to the volumes. Let D
be the normal vapour density, corresponding with the undissociated
substance, A the observed vapour density, then :

£ = A[1 -f y(*- 1)]

Z>-A
= A (x - 1)

If d is the vapour density con-responding with complete dis-
sociation, d = D/x.

In the case of phosphorus pentachloride, x — 2, hence :

D- A

«y 7^—

Thus d == J/>, i.e., on complete dissociation the vapour density
has half the normal value.

The dissociation of PC15 is easily demonstrated. Both PC16 and
PC13 are colourless in the form of vapour ; C12 is greenish-yellow.
The vapour of PC15, however, also shows a greenish -yellow colour,
which becomes deeper as the temperature increases. At the same time
the density (reduced to S.T.P.) decreases. Hence as the proportion

154 INORGANIC CHEMISTRY CHAP.

of chlorine, or the extent of dissociation, increases, so the density
decreases, under these conditions. The vapour also turns potassium
iodide and starch paper blue, indicating the presence of chlorine.

At 200° and 1 atni. pressure, the vapour density of phosphorus
pentachloride is 6741. The density corresponding with no dis-
sociation is iPC!5 = 100-1. Thus, Z> = 100-1, A = 6741.

.'. y = - - — 0485. Thus, out of every 100 molecules

of PC15 heated to 200° under 1 atm. pressure, 48-5 are dissociated
into PG13 -+- C12. The vapour densities and dissociations at various
temperatures (1 atm. press.) are (Cahours, 1847) :

£<> 182 190 200 230 250 274 288 300 336°
A 70-5 69-4 67-4 59-7 55-6 53-4 51-0 50-7 50-8
y 0-417 0-443 0-485 0-674 0-800 0-875 0-962 0-973 0-970

If we plot y and A against t, we obtain the dissociation curves,
Fig. 79. These show three parts : two natter end parts, near the
limiting values of the densities corresponding with no dissociation
and complete dissociation, respectively, and a rising or falling
intermediate portion, where the influence of temperature is marked.

The mechanism of chemical reactions. — If the elements chlorine
and hydrogen are brought together they react to form the com-
pound hydrochloric acid ; this is called combination. By suitable
means (e.g., electrolysis) it is possible to recover from hydrochloric
acid, qualitatively and quantitatively, the elements of which it is
composed, and the process is called decomposition.

Dalton regarded these changes as real combinations between
atoms, and decompositions of compounds into atoms : H 4- Cl = HC1,
and HC1 — H -|- Cl, so that a nomenclature originally applied to
substances was appropriate also to the atoms. With the advent
of the molecular theory, this point of view could not be main-
tained. The reactions had then to be formulated as follows :

(1) H2 + C12 = 2HC1, or HH + C1C1 = HC1 + HC1, and

(2) 2HC1 - H2 -f C12, or HC1 + HC1 = HH + C1C1.

They are now seen to be examples, not of simple combination
and decomposition, but of double decomposition, i.e., a special case
of rearrangement of the atoms in different molecules, when the
numbers of molecules before and after the reaction are the same.

In the same way, cases of displacement often lead to the elimination
of molecules, not of atoms : Hg012 + Zn = ZnCl2 + Hg (atom) ;
Zn + 2HC1 = ZnCl2 + H2 (molecule).

Cases of true combination between atoms alone, or decomposition
into atoms alone, are rare. Combination occurs between atoms
and molecules, e.g., Hg -f C12 — HgCl2, or between molecules and

AVOGADRO'S HYPOTHESIS AND THE MOLECULE

155

molecules, e.g., CO + C12 = COC12, but is rarely observed between
atoms alone : 1 + 1 = 1%. Many apparent cases of combination are
really examples of double decomposition : HH +11 = HI + HI.
Again, a compound decomposes into molecules, e.g., CaCO3 = CaO +
CO 2, or into molecules and atoms, e.g., 2HgO = 2Hg + O2, but
seldom into atoms alone. Other apparent cases of decomposi-
tion are really double decompositions : HI + HI = HH +11, or

0-8

0-6

0-4

0-2

150

200 250 30O

FIG. 79. — Dissociation Curves.

850*0.

2HI = H2 + I2. Double decomposition, in fact, is the commonest
type of chemical change.

The mechanism of chemical changes then becomes much more
complicated than on the basis of the atomic theory alone. Thus,
the formation of water from gaseous oxygen and hydrogen, instead
of being a simple combination : 2H + O = H2O, occurs between
molecules, and possibly in stages, various types of which are
possible :

1. Decomposition of the molecules into atoms, followed by simple

156 INORGANIC CHEMISTRY CHAP.

combinations between the latter : H2 = 2H, O2 = 2O, and 2H -f- O
= H20.

2. Direct combination between molecules, to form hydrogen peroxide,
H2O2 ; H2 + O2 = H2O2, followed by

(i) : decomposition of the hydrogen peroxide by heat :

2H2O2 = 2H2O + O2, or H2O2 = H2O + O, followed by H2 + O =
H2O; or, (ii) : reduction of the hydrogen peroxide by another
hydrogen molecule : H2O2 -f H2 = 2H2O.

3. Double decomposition between hydrogen and oxygen molecules,
to form a water molecule and an oxygen atom, the latter combining
with another hydrogen molecule to form water : H2 + O2 = H2O -f- O
and H2 + O = H20.

Traube favoured scheme (2) ; Dixon's experiments led him to the
opinion that scheme (3) is the most likely; scheme (1) has few sup-
porters.

In the present state of chemistry it cannot be said with certainty
which, if any, of these alternative groups of reactions really goes on
in the combustion of hydrogen, or whether two or more of them
proceed simultaneously. The case is no better with other simple
reactions, and a large and intensely interesting field of inquiry
still awaits investigation.

The case of isomeric change is also considerably amplified by
the molecular theory. Two possibilities are obvious : (1) the
different substances of the same empirical formula have the same,
molecular weight', they are then called metamers, and their inter-
conversion, metameric change ; or (2) they have different molecular
weights, when those of higher molecular weight are called polymers,
and their formation from the substance of lower molecular weight
is called polymerisation.

SUMMARY OF CHAPTER IX

Gay-Lussac's law of volumes : when chemical changes occur between
gases, there is always a simple relation between the volumes of the inter-
acting gases, and also of the products if these are gaseous.

Avogadro's hypothesis : this explains Gay-Lussac's law, and states
that : equal volumes of all gases and vapours, under the same conditions
of temperature and pressure, contain identical numbers of molecules.
It applies exactly to gases under very low pressures ; under ordinary
conditions it is only an approximate law.

A molecule is the smallest portion of a substance which can exist in
the free state.

An atom of an element is the smallest portion of it which can exist
in a molecule of a compound. In some cases (e.g., Hg) the atom is
identical with the molecule, but more usually the molecule consists of
two or more atoms.

ix AVOGADRO'S HYPOTHESIS AND THE MOLECULE 157

The molecular weight of a substance is the ratio of the weight of a
molecule of that substance to the weight of an atom of hydrogen. It
is twice the relative density of the gaseous or vapour form of the
substance (H = 1), since the hydrogen molecule contains two atoms,
H2.

The molecular weight in grams of any gas (gm. mol.) occupies at
S.T.P. a volume of 22-24 litres (gm. mol. vol.).

Many compounds on heating dissociate, i.e., are partially decomposed,
to an extent increasing with the temperature, in such a way that the
products recombine on cooling. If change of density occurs, the
degree of dissociation, i.e., the fraction of the total number of mole-
cules which are broken up, may be calculated from the equation :
y = (D — A)/ A (a; — 1), where D = density of undissociated
substance ; A = observed density, both reduced to S.T.P., and
x = number of molecules formed on dissociation from one molecule of
the substance.

EXERCISES ON CHAPTER IX

1. State Gay-Lussac's law of volumes, and describe two experiments
which could be performed to demonstrate its truth. Does the law
hold accurately ? What explanation may be given for the deviations
found by experiment ?

2. Describe the evidence which led Avogadro to assume that the
smallest particles of gases . usually consist of more than one atom.
What reason is there for the assumption that the molecules of hydrogen,
oxygen, and chlorine consist of two atoms, whilst that of mercury vapour
consists of only one ?

3. What weight of barium peroxide must be decomposed by heating
to give 42 litres of oxygen at 18° and 740 mm. ?

4. The molecular weight of cyanogen is 52-08 (O = 16). Find its
density referred to air = 1, and its normal density. It contains 46-08
per cent, of carbon and 53-92 per cent, of nitrogen ; what is its formula ?
What volumes of nitrogen and carbon monoxide would be formed by
exploding 1 litre of cyanogen with oxygen ?

5. Explain why the atomic weights of carbon, oxygen, and sulphur
are taken as 12, 16, and 32 instead of 6, 8, and 16, the values adopted
by Gmelin.

6. The normal densities of chlorine, carbon dioxide, and ammonia
are 3-220, 1-9768, and 0-7708 gm. per litre, respectively. Calculate the
gram-molecular volumes (H = 1), and explain why these are not
exactly equal to 22-24 litres.

7. Show that the molecular weight in ounces of a gas occupies nearly
\ he same volume in cubic feet as the molecular weight in grams occupies
in litres.

8. Calculate the atomic weights and molecular volumes of chlorine
and hydrogen on the atomic weight standard O = 100.

9. Discuss the nature of chemical change from the point of view of the
molecular theory. Criticise the statement : " hydrogen and oxygen
combine to form water." In what sense is it correct ?

10. Explain, with examples, the methods used in deciding which
multiple of the equivalent is the atomic weight of an element.

158 INORGANIC CHEMISTRY CH. ix

The chloride of an element contains 37-322 per cent, of chlorine. The
vapour density of the chloride is 190 (H = 1). The specific heat of
the element is O0276. Find the atomic weight of the element, and the
formula of the chloride.

11. Describe the cases of abnormal vapour densities met with. What
explanation of these has been given, and what evidence is there of its
correctness ?

12. Define dissociation. In what way does it differ from such changes
as the decomposition of potassium chlorate by heat ?

13. Using a porcelain Victor Meyer apparatus (p. 87) the following
data were obtained for iodine : 0-0874 gm. of iodine displaced 13-7 c.c.
of air. Barometer 722-8 mm. : temperature of room 21-5°; vapour
pressure of water at 21-5° = 19-2 mm. Calculate the vapour density
of iodine, and the degree of dissociation at the temperature of the
experiment (I = 127).

14. Under what conditions is Avogadro's law strictly applicable ?
Show how the molecular weight of a gas may be found accurately from
the density. The weight of 1 litre of a gas at S.T.P. is 1 -2507 gm. ; its
compressibility coefficient is — 0-000559. Find its molecular weight
(H = 1).

CHAPTER X

OXYGEN. (0 = 15-87)

Occurrence of oxygen. — The element oxygen (O = 15-87) occurs
in the free state as a gas, of the molecular formula O2. It is colourless,
odourless, 'and tasteless, and supports combustion and respiration.
It is the uncombined oxygen in the atmosphere, where it occurs to
the extent of 21 per cent, by volume or 23 per cent, by weight,
which takes part in processes of combustion ; its functions in respira-
tion make it the most important element from the biological point
of view. Oxygen is sparingly soluble in (and may therefore be
collected over) water, but the small quantity of oxygen dissolved
in river and sea waters is essential to the life of fish.

Combined oxygen occurs in water, in vegetable and animal tissues,
and in nearly all minerals and rocks. Oxygen occurs to a larger
extent in the earth's crust than any other element ; it makes up
about 50 per cent, of the total quantity of the terrestrial elements.

Oxygen was first isolated by Scheele in 1772, and was discovered
independ^ , by Priestley in 1774 (p. 44).

Accor< vorth, the Chinese philosopher, Mao Khoa (eighth

century' Aary elements : Yin (the weak), and Yang

(the strong). . and yang are combined with fire. When

charcoal is bi . atr^J^tig is left, whilst yin could be obtained by

heating a subs.. .'•'•• -inil (possibly nitre). The Greek alchemist

Zosimus (third centiny) alao refers to a gaseous body evolved on heating
a substance floating onv'the surface of heated mercury (possibly
mercury oxide).

Preparation of oxygen. — Oxygen may be obtained by simply
heating certain metallic oxides, viz., those of mercury, silver, gold,
and the platinum metals.

If mercuric oxide is heated in a hard glass tube it decomposes ;
globules of mercury collect in the cooler parts of the tube ; oxygen gas
is evolved, and may be collected over water : 2HgO = 2Hg -j- O2
(Fig. 24). Oxide of silver, precipitated from silver nitrate solution

159

160 INORGANIC CHEMISTRY CHAP.

by caustic potash (in absence of carbon dioxide), gives very pure oxygen
when heated : 2Ag2O = 4Ag + O2.

Oxygen is not evolved on heating the lower oxides of metals other
than the above, but many higher oxides, including peroxides (p. 134),
lose a portion of their oxygen at more or less elevated temperatures.
Examples are hydrogen peroxide, H202 ; barium peroxide Ba02 ;
lead dioxide, Pb02 ; manganese dioxide, MnO2.

2H202 = 2H2O + O2 ; 2Ba02 = 2BaO (baryta) + 02 ; 2PbO2 =
2PbO -f02 ; 3Mn02 = Mn304 (trimanganic tetroxide) + O2.

Manganese dioxide (pyrolusite) is decomposed on heating to bright
redness in an iron tube, and this reaction was formerly a cheap method
of preparing oxygen on a moderate scale. The dioxide evolves oxygen
at a lower temperature when heated with concentrated sulphuric acid
in a glass flask : 2MnO2 + 2H2SO4 = 2MnSO4 (manganous sulphate)
+ 2H2O + Oa : frothing, however, occurs, and dangerous explosions
result if water is drawn back into the heated acid from the pneumatic
trough. Lead dioxide loses oxygen fairly readily at a dull red heat,
but the lead monoxide, or litharge, PbO, readily attacks glass or
porcelain. None of these methods is now used.

Oxygen may be obtained from water by electrolysis (p. 56), or by
removing the hydrogen with chlorine ; the latter readily combines
with hydrogen to form the stable hydrochloric acid, HC1, but does
not unite directly with oxygen : 2H20 + C12 = 4HC1 + O2.

EXPT. 61. — A stream of chlorine, generated from potassium per-
manganate and concentrated hydrochloric acid in a flask, is passed
through water boiling in a second flask, and the gas is then passed through
a silica tube packed with bits of broken porcelain and heated to bright
redness in a furnace (Fig. 80). The gas is passed through caustic soda
solution in a wash-bottle to remove excess of chlorine, and hydro-
chloric acid, and the oxygen is collected over water.

All modern processes for the preparation of oxygen in the
laboratory make use of salts rich in oxygen : chlorates (e.g., KC103),
bromates, iodates, nitrates, dichromates (e.g., K2Cr207), and per-
manganates (e.g., KMnO4).

The production of oxygen by heating nitre, when potassium nitrite
is left as a residue, has already been described (p. 43) : 2KNO3 =
2KNO2 + O2. The method is not used in the preparation of oxygen,
as a high temperature is required.

Potassium chlorate, KC103, is, the most convenient source of
oxygen in the laboratory. The crystals, which are anhydrous, melt
at 372°, and on heating to 380° in a hard glass flask bubbles of oxygen
are evolved : (1)2KC1O3 = 2KC1 + 302. As the reaction proceeds,

OXYGEN

161

the evolution of oxygen slackens, and the salt becomes pasty,
finally almost solid, although decomposition is not nearly complete,
i.e., the residue is not wholly potassium chloride, KC1.

At this stage of the reaction the residue contains potassium
chloride and potassium perchlorate, KC104, a salt richer in oxygen
than the chlorate, which is produced by the reaction : (2) 4KC103 =
3KC104 -f KC1. The KC1 and KC104 may be separated by treat-
ment of the cooled residue with cold water, when the former
salt dissolves. (CJ. p. 372.) If the temperature is raised when
the salt becomes pasty, the mass fuses again, oxygen is evolved,

FIG. 80. — Decomposition of Steam by Chlorine.

and finally, when all has become solid, potassium chloride is left :
(3) KC1O4 = KC1 + 202. Reactions (1) .and (2) proceed simul-
taneously and independently from the commencement.

At high temperatures another mode of decomposition : (4) 4KC1O3
= 2K2O + 2Cla + 5O2, takes place to a slight extent, the gas containing
a little chlorine, and showing a slight fog, due to suspended solid
potassium oxide, K2O. A mixture of potassium and sodium chlorates
liberates oxygen at a lower temperature than potassium chlorate alone.

Potassium permanganate on heating to 240 ° in a glass tube evolves
very pure oxygen, leaving a black powdery residue of potassium
manganate, K2Mn04, and manganese dioxide : 2KMnO4 = K2MnO4
+ Mn02 + O2. By adding a little water to the residue, a dark
green solution of the manganate is formed.

Many other oxy-compounds may be used as sources of oxygen gas.
Thus, if a solution, or paste, of bleaching powder or chloride of lime,

162 INORGANIC CHEMISTRY CHAP.

containing the compound Ca02Cl2, is heated to 75° with a few drops
• of cobalt or nickel chloride solution, oxygen is rapidly evolved
(Mitscherlich, 1843) :

Ca02Cl2 = CaCl2 + O2.

If a little manganous sulphate is added, the oxygen is free from
chlorine. A solution of bleaching powder alone decomposes only
slowly. Cobalt or nickel oxides are precipitated by the free lime
contained in the bleaching powder : CoCl2 + Ca(OH)2 = CoO +
CaCl2 -j- H20. A higher, unstable oxide, Co203 or CoO2, appears
to be alternately formed and reduced, thus acting as a carrier of
oxygen : 2CoO + Ca02Cl2 = 2Co02 + CaCl2 = 2CoO + 02 + CaCl2
(Fleitmann, 1865). A mixture of copper and ferrous sulphates,
neither of which alone is active, accelerates the decomposition of
bleaching powder solution (Jaubert). The same reaction occurs if
chlorine gas is passed into boiling caustic soda solution, or milk
of lime, to which a few drops of cobalt or nickel chloride have been
added : 4NaOH + 2C12 = 4NaCl + 2H20 +02.

EXPT. 62. — Add a little nickel chloride solution to a solution of
caustic soda : a light green precipitate of the hydrated lower oxide,
NiO,H2O, is thrown down. Pour a little of the suspension of this into
bleaching powder solution. The precipitate at once becomes oxidised
to a black substance, Ni2O3,a;H2O, and oxygen is freely evolved on
warming.

Chromium trioxide and potassium dichromate evolve oxygen when
heated in a flask with concentrated sulphuric acid, the red colour
of these compounds changing to the dark green colour of
chromium sulphate, Cra(S04)3 :

K2Cr2O7 + H2S04 = K2S04 + 2Cr03 + H2O
2Cr03 + 3H2SD4 = Cr2(S04)3 + 3H20.

Chromium trioxide also decomposes when heated alone, although
a little sublimes unchanged : 4Cr03 = 2Cr203 (green) -|- 302.

If the residue left after decomposing potassium dichromate with
sulphuric acid is cooled, diluted with an equal volume of water, and
allowed to stand for some time in a loosely-covered beaker, beautiful
deep-violet octahedral crystals of chrome alum, K2SO4,Cr2(SO4)3,24H2O,
separate out.

Potassium permanganate explodes violently when warmed with
concentrated sulphuric acid, but readily yields very pure oxygen
if ordinary hydrogen peroxide (4 per cent, solution) is mixed with
a solution of the permanganate acidified with dilute sulphuric
acid : the two highly oxidised compounds mutually decompose
each other, yielding a nearly colourless solution :

2KMn04+ 3H2S04 + 5H2O2 = K2S04 + 2MnS04 + 8H20 + 5O2.

OXYGEN

163

KXPT. 63. — A solution of 5 gm. of KMriO4 in a cooled mixture of
100 c.c. of water and 50 c.c. of concentrated sulphuric acid is dropped
from a tap-funnei into 100 c.c. of " 10 volumes " hydrogen peroxide
in a flask (Fig. 81). The evolved oxygen is collected over water.

The preparation of oxygen in the laboratory.— The evolution of
oxygen from potassium chlorate is greatly accelerated if manganese
dioxide is mixed with the salt.

EXPT. 64. — Fuse a little potassium chlorate in a test-tube, and
keep the temperature below the point at which oxygen is evolved.
Now add a little powdered manganese dioxide : a rapid evolution
of oxygen occurs.

If enough manganese dioxide is ground in a mortar with potass-
ium chlorate to render the mixture black, and
this oxygen mixture is heated in a glass tube or
flask, decomposition occurs rapidly at a tem-
perature below the melting point of the chlorate,
oxygen being freely evolved. The heating must
be carefully regulated, as the decomposition of
potassium chlorate, unlike that of mercuric oxide
(p. 24), evolves heat, and under certain conditions
may become explosive.

The manganese dioxide undergoes no per-
manent chemical change in the reaction : it may
be completely recovered by washing out the potass-
ium chloride from the residue with water. The
oxygen prepared in this way contains a little
chlorine, which may be removed by washing with
caustic soda solution ; the gas is often misty
from suspended particles of potassium chloride
or hydroxide, the latter formed from water
and potassium oxide (equation (4) on p. 161).
These subside on standing over water.

EXPT. 65. — Mix 25 gm. of powdered potassium chlorate with a
few grams of powdered manganese dioxide in a mortar. Place the
mixture in a wide test-tube, and tap the tube so as to leave a free
passage for the gas from the bottom of the tube. Fit the tube, with
a good cork and a wide (^-in.) glass delivery tube, to a Woulfe's bottle
containing caustic soda solution, as shown in Fig. 82. The caustic
soda removes any trace of chlorine from the gas. Support the test-
tube in a horizontal position in a clamp, and heat the mixture
gently with a slightly luminous flame, beginning at the end near the
cork and moving towards the closed end as the reaction proceeds.
If the evolution of gas becomes violent, withdraw the flame till it

M 2

FIG. 81.
Preparation of
Oxygen from

Potassium

Permanganate

and Hydrogen

Peroxide.

164

INORGANIC CHEMISTRY

slackens. The gas may be collected in jars over water, or in a metal
Pepys' gas-holder, as shown. The latter stands in a trough of water,
and the delivery tube is inserted into the lower opening. When the
gas has been collected, this opening is closed by a screw stopper. The
funnel tube, A, and gas-holder are filled with water before the collection
of the gas. When the gas is no longer evolved, the test-tube is
taken off to prevent liquid being drawn back into the tube and cracking
it. Jars may be filled in the upper trough of the gas-holder over the
short tube, B, by opening the taps on A and B.

Warning. — Manganese dioxide has sometimes been adulterated with
powdered coal ; it then explodes violently
on heating with chlorate. More than one
death has been caused in this way, and a
little of the mixture should always be heated
in an open test-tube before beginning the
experiment, in order to be sure that no
(ration occurs.

Other oxides, such as ferric oxide, cupric
oxide, and chromium oxide, act similarly
to manganese dioxide : they are also left
unchanged after the reaction. This action
of manganese dioxide, discovered by
Dobereiner in
1832, is an
example of
numerous re-
acti ons in
which a sub-
stance accele-
rates a chemi-
cal change

FIG. 82. — Preparation of Oxygen from Potassium Chlorate aod
Manganese Dioxide.

Such

without itself, apparently, taking part in the reaction,
substances were called catalysts by Berzelius (1835).

Combustion. — The combination of substances with oxygen,
when attended with the evolution of heat and light, is called
combustion. Substances which burn in air do so with greatly
enhanced brilliancy in pure oxygen, since the nitrogen in air acts
as a diluent, absorbing part of the heat given off in the combustion.

The combustion of sulphur, phosphorus, and carbon, giving acidic
oxides, has already been described (p. 49) :

S + O2 = SO2 ; SO2 -f H2O = H2SO3 (sulphurous acid).

2S -f 3O2 = 2SO3 ; SO3 + H2O = H2SO4 (sulphuric acid)

4P + 5O2 = 2P2O6 ; P2O5 -f- H2O = 2HPO3 (metaphosphoric acid).

C -h O2 = CO2 ; CO2 + H2O = H2CO3 (unstable carbonic acid).

OXYGEN

165

The substances are conveniently burnt in globes of oxygen inverted
over upright deflagrating spoons (Fig. 83).

Magnesium ribbon, if ignited in air and inserted into a jar of oxygen,
burns with a blinding white light, forming white solid magnesium oxide,
MgO, which is a weakly basic oxide, and turns red litmus paper blue
when moistened and laid upon it. Sodium and potassium, when heated
in iron deflagrating spoons until they begin to burn, and then lowered
into dry jars of oxygen, burn with bright yellow and purple flames,
respectively, forming orange-yellow solid oxides which dissolve in
water with evolution of oxygen and
formation of strongly alkaline sodium
and potassium hydroxides :

2Na + O2 == Na2O2 (sodium per-
oxide) ; 2Na2O2 + 2H2O = 4NaOH
(caustic soda) -f- O2 ;

2K + 2O2 -- -- K2O4 (potassium
tetroxide) ; 2K2O4 + 2H2O = 4KOH
(caustic potash) + 3O2.

A spiral of iron wire, tipped with
a bit of .burning wood, burns bril-
liantly, giving off a shower of bright
sparks, when lowered into a bottle
of oxygen. Black oxide of iron,
Fe3O4, is formed in fused globules,
which crack the bottle when they fall
on it, even if water is poured into the
bottle before the experiment.

A jet of hydrogen burns in a jar of
dry oxygen, producing wa.ter, which
condenses in drops on the cold sides
of the jar : 2H2 + O2 = 2H2O. If
a jet of oxygen is thrust into an

inverted jar of hydrogen, burning at the mouth, the oxygen takes fire,
and continues to burn in the atmosphere of hydrogen (Fig. 84). The
terms combustible, and supporter of combustion, are, therefore, purely
relative.

EXPT. 66. — Dry barium or strontium chlorate is heated in a vertical
spoon until it evolves oxygen freely. A globe of coal gas is then lowered
over the spoon into water in the trough (Fig. 83). The oxygen from the
chlorate, if the latter is sufficiently heated, takes fire, and burns in the
coal gas, the flame being coloured intensely green or crimson by the
volatile barium or strontium compounds, respectively.

Many combustible substances, in a finely divided condition,
ignite spontaneously in air or oxygen.

FIG. 83. — Apparatus for Combustions
in Oxygen.

166 INORGANIC CHEMISTRY CHAP.

EXPT. 67. — By means of a brush trace letters on a piece of filter-
paper with a solution of phosphorus in carbon disulphide. When the
solvent evaporates, the finely divided phosphorus ignites, leaving the
charred letters on the paper.

EXPT. 68. — Precipitate a solution of lead acetate with a solution of
Rochelle salt, KNaC4H4O6. The white precipitate of lead tartrate,
PbC4H4O6, is filtered, washed, and dried in a steam-oven. Small portions
are placed in narrow tubes, sealed at one end and drawn out at the other.
The tartrate is heated until fumes are no longer evolved, and the tubes
are sealed. If a tube, after cooling, is cut with a file, and the finely divided
lead shaken out, the metal glows brightly, forming yellow fumes of lead
oxide, PbO. This form of the metal is called pyorophoric lead.

Many substances, such as phosphorus, oxidise slowly when
exposed to air or oxygen, without catching fire,
because the heat produced is dissipated too
rapidly to raise the mass to the ignition point.
Oily cotton-waste, however, may become heated
to the ignition point if stored in heaps exposed
to air. This slow process of oxidation is known
as autoxidation.

Oxygen is absorbed from gaseous mixtures by : (i) a
solution of pyrogallol in caustic potash, which turns
blacl5*";( 1 60 grams of KOH, 10 grams of pyrogallol,
130 c.c. of water) ; (ii) moist phosphorus (this does
not glow in pure oxygen) ; (iii) an acid solution of
Fia. 84. chromous chloride, CrCl2, which turns from blue to

0xygHnydrog?nng ^ g1"6611' owing to the formation of chromic chloride :
4CrCl2 + O2 4- 4HC1 = 4CrCl3 + 2H2O ; (iv) by mixing

the gas with excess of hydrogen, and passing over platinum black at

100°, or platinised asbestos at a dull red heat, when water is formed ;

one -third of the contraction of the gas then represents the oxygen

contained in it : 2H2 + O2 = 2H2O (liquid).

Catalysis. — The action of manganese dioxide, copper oxide,
and ferric oxide in promoting the decomposition of potassium
chlorate by heat, and the similar effect of cobalt and nickel oxides
on bleaching powder, have been described. These substances
appear to act by contact, hence their effect was called contact
action by Mitscherlich ; the usual name, due to Berzelius, is
catalytic action or catalysis. The manganese dioxide is called a
catalyst.

A catalyst is a substance which alters the speed of a chemical
reaction without itself undergoing permanent chemical change ;
in most cases it accelerates the reaction, when it is called simply

x OXYGEN 167

a catalyst, but in some cases it retards it, when it is called a
negative catalyst. It is essential that a true catalyst shall undergo1
no permanent chemical change ; it must be left after the reaction
of the same chemical composition as at the beginning, but not
necessarily in the same physical state. Very small quantities of
a catalyst will therefore serve to bring about the decomposition,
or other chemical change, of large quantities of materials. The
importance of catalysts in chemical industry is therefore clear.

The first reasonable theory of catalytic action was due to J. Mercer
(1842). This assumed that the catalyst forms with one of the
final products of reaction an unstable intermediate compound,
which then breaks up, reproducing the catalyst in its original
chemical composition, and liberating the product of reaction.
This series of alternating, or cyclic reactions, so called because the
catalyst goes through a series of complete cycles of changes and
returns to its original state after each, is regarded by this theory
as the cause of catalytic action. Thus, manganese dioxide in
presence of a powerful oxidising agent, such as potassium chlorate,
tends to pass into a higher stage of oxidation, say Mn207, which
would give potassium permanganate with the potassium salt.
At the high temperature, however, this higher oxide can hold its
oxygen only transiently ; it breaks up, giving gaseous oxygen,
and forming manganese dioxide again : ' £

KC103 + 2Mn02 -> KC1 + Mn2O7 -> KC1 + 2Mn02 + 30.

Fowler and Grant (1890) showed that only oxides which can form
unstable higher oxides, again decomposed by heat, can act cata-
lytically in the decomposition of potassium chlorate. Thus,
Mn02 -> Mn03 or Mn207 ; Cr203 -> Cr03 ; Fe203 -> Fe03 ; all
these higher oxides are known in the form of salts : K20,Mn03
(manganate) ; K2O,Mn207 (permanganate) ; K20,Cr03 (chromate) ;
K20,Fe03 (ferrate). Copper oxide probably forms an imperfectly
known higher oxide (? Cu02). Oxides which do not form higher
oxides, such as zinc oxide or magnesium oxide, act only very
feebly (to the same extent as powdered glass), whilst acidic oxides,
such as alumina, A1203, vanadium pentoxide, V205, or tungsten
trioxide, W03, give both chlorine and oxygen : 2KC1O3 (or
K20,C1205) + W03-> K20,W03 + Cl205-> K2O,W03 + C12 + 5O.

EXPT. 69. — Fuse some potassium chlorate in two hard glass tubes.
To one add a very small quantity of manganese dioxide, to the other
a very small quantity of chromium sesquioxide, Cr2O3. Observe that
(i) oxygen is evolved ; (ii) the fused salt becomes permanently pink
(KMnO4), and yellow (K2CrO4), respectively. KMnO4 cannot exist
alone at the temperature of the fused chlorate, hence it must be con-

168 INORGANIC CHEMISTRY CHAP.

tinuously decomposed and reproduced by a series of cyclic actions such
as that described above. A little ferric oxide, Fe2O3, produces a violent
effervescence, and on cooling the mass is slightly pink, from the forma-
tion of ferrate, K2FeO4.

McLeod (1889) observed that pieces of manganese dioxide put
into fused chlorate break up into a very fine powder. The physical
state of the manganese dioxide changes, which suggests that it
has entered into reaction and been reproduced. Traces of chlorine
are always evolved in the preparation of oxygen from chlorate,
and McLeod suggested that chlorine and potassium permanganate
are intermediate products in the decomposition :

(1) 2KC1O3 + 2Mn02 = 2KMn04 + C12 + O2.

(2) 2KMn04 = K2Mn04 + Mn02 + 02.

(3) K2Mn04 + C12 = 2KC1 + Mn02 + 02.

If chlorine escapes, however, the residue should contain mangan-
ate ; this is never found, so that probably the chlorine is produced
by a secondary reaction : 4KC1O3 = 2K2O + 2C12 + 502, which
is known to take place at 360°. Reactions (2) and (3) are also
known to take place, but it is doubtful if (1J occurs. This reaction,
however, is the basis of McLeod 's scheme. No perchlorate is formed.

It may be difficult to see how manganese dioxide can exert any action
on solid chlorate, since the catalytic effect occurs below the fusion point
of the latter. But some local fusion probably occurs on account of the
heat evolved in the reaction (flashes of light are always seen), and in any
case L. H. Parker (1914-18) has shown that chemical action may occur
between solids. Thus, he found that the reaction :

BaCO3 -f Na2SO4 = BaSO4 -f Na2CO3,

and the reverse reaction, take place to a limited extent when the
dry powdered mixture is heated short of fusion, or simply triturated
in a dry mortar. Reaction also occurs in the dry powder when it is
strongly compressed, as was shown by Spring.

Manufacture of oxygen. — On the large scale oxygen is prepared
(1) from water, by electrolysis (p. 56) ; (2) from air. In the
preparation from air two kinds of processes are used : (a) physical
methods, (b) chemical methods.

All the chemical methods depend on the use of a substance, A,
which takes up oxygen from the air under certain conditions,
leaving the nitrogen : (i) A + O2 -f- 4N2 = AO2 + 4N2. Under
other conditions the compound AO2 can be broken up again into
A, which is used over again, and oxygen : (ii) AO2 — A + O2.
Reactions (i) and (ii) alternate.

Boussingault in 1852 noticed that if baryta, BaO (which is a
substance similar to quicklime, CaO), is heated in a porcelain tube

x OXYGEN 169

to dull redness, it can absorb oxygen from air passed over, giving
barium peroxide : (i) 2BaO -f- 02 = 2BaO2, whilst the nitrogen
is not absorbed. If the barium peroxide is now heated to bright
redness, it gives off oxygen, leaving baryta : (ii) 2Ba02 = 2BaO + 02.
The reaction is therefore reversible, and proceeds in one direction
or the other according to the temperature : 2Ba02 — - 2BaO -f O2.
It was found that the baryta rapidly became inactive ; carbon
dioxide in the air produced barium carbonate, BaCO3, which
is only decomposed at a white heat, and the silica from the
tube formed barium silicate, BaSiO3. Both these substances
cover the baryta. By using purified air, and iron retorts, the
brothers Brin in 1879 succeeded in keeping the baryta active.
They found that the process could be worked at one temperature
if, during the absorption, the air was under 2 atm. pressure, whilst
the peroxide was decomposed on reducing the pressure to about
2 in. of mercury. The iron retorts were placed vertically in a
furnace heated by gas to about 700°. This Brin process was the
principal technical method until 1902 ; it has now given way to
the liquid air process (p. 175).

Tessie du Motay in 1866 passed air over a mixture of caustic soda and
manganese dioxide heated to dull redness in retorts. Sodium manganate
was produced :

(i) 2MnO2 + 4NaOH + O2 (air) = 2Na2MnO4 + 2H2O.
The temperature was then raised to a bright red heat, and steam passed
over the manganate, when oxygen was liberated :

(ii) 2Na2MnO4 + 2H2O = 2MnO2 + 4NaOH -f O2.
The temperature was allowed to fall, and air passed over the residue ;
manganate was again formed. Reactions (i) and (ii) thus alternated.
The process was at one time used in Paris, but has been entirely
abandoned.

Kassner in 1889 heated a mixture of litharge and chalk in air at 600°.
This gave calcium plumbate, 2CaO, PbO2, or Ca2PbO4 :

(i) 4CaCO3 + 2PbO + O2 (air) = 2Ca2PbO4 + 4CO2.
Moist furnace gas was passed over the plumbate at 80-100° :

(ii) Ca2PbO4 + 2CO2 = 2CaCO3 + PbO2.

On heating to 500°, the PbO2 was decomposed, with evolution of
oxygen, leaving PbO and CaCO3. This process is complicated, and
is not used.

A few physical processes were proposed before the present in-
dustrial method was adopted. Graham found that oxygen passes
through an unvulcanised rubber membrane two and a half times
as fast as nitrogen, and by pumping air through a rubber bag

170 INORGANIC CHEMISTBY CHAP.

by means of a mercury pump he obtained a gas containing 42 per cent,
of oxygen, which rekindled a glowing chip. This process, depending
on the selective permeability of a membrane, is called dialysis.
Again, if air is shaken with water, oxygen is dissolved more readily
than, nitrogen, and the gas liberated on heating or reducing the
pressure is richer in oxygen (p. 97). By working under pressure,
and repeating the process four or five times. Mallet obtained a gas
containing over 75 per cent, of oxygen.

If air is slowly passed through the stem of a clay tobacco-pipe
enclosed in a partially exhausted glass tube, the lighter nitrogen
diffuses through the porous tube more rapidly than the oxygen,
in the inverse ratio of the square roots of the densities (Graham) :

Speed of diffusion of nitrogen V 16 1-07
Speed of diffusion of oxygen V 14 ~~ 1*00

The issuing gas is therefore richer in oxygen than air. This
process was called atmolysis by Graham.

Since oxygen is slightly heavier than nitrogen, Mazza ( 1901 ) attempted
to separate air into the two gases by passing it through a centri-
fugal sieve : needless to say, the method failed.

The only process now used for the manufacture of oxygen is
the fractional distillation of liquid air.

Liquefaction of gases. — Sulphur dioxide was liquefied by cooling
and pressure by Monge and Clouet ; in 1805 chlorine and ammonia
were reduced to the liquid state by Northmore. In 1823 liquid
chlorine was again obtained by Faraday, by warming chlorine
hydrate in one limb of a sealed A-tube, the other limb of which
was cooled in a freezing mixture. In later experiments, Faraday
was able to liquefy hydrogen sulphide, hydrogen chloride, carbon
dioxide, nitrous oxide, cyanogen, and ammonia ; but oxygen,
nitrogen, and hydrogen resisted all attempts to reduce them to
the liquid state.

Most of the attempts relied on the application of pressure to the
gases. Some gases may be liquefied by the application of pressure
without very strong cooling : in the following table the pressures
in atm. required to liquefy the gases at 0° are given : —

Sulphur dioxide 1-54 Ammonia 4-19

Chlorine ... 3-66 Carbon dioxide 39'0 (at 15°)

The application of pressures up to 2000 atm., however, was tried
by Natterer in the case of the gases nitrogen, oxygen, and hydrogen,
without result.

In 1869 Andrews discovered that a gas cannot be liquefied by
any pressure, however high, unless it is previously cooled below
what is called the critical temperature of the gas. Just below

OXYGEN

171

this temperature the gas is liquefied by the application of what
is known as the critical pressure. The volume occupied by 1 gm. of
a substance at the critical temperature and under the critical
pressure is called the critical volume.

The critical temperatures of the so-called permanent gases lie
below the lowest temperatures attained by the older experimenters.
As soon as it was clear that strong cooling was necessary in the
case of these gases, and that high pressures alone could never
succeed, the problem was solved, independently, by Pictet and
Cailletet in 1879.

Pictet used the apparatus shown in Fig. 85. Oxygen was gener-
ated in the retort P by heating potassium chlorate, and was
compressed by its
formation in the
copper tube 0,
fitted with a pres-
sure gauge Q and
release valve N,
cooled in liquid
carbon dioxide L
boiling under re-
duced pressure. This
carbon dioxide was
again liquefied by
the pump G in a
second copper tube,
EF surrounded by
liquid sulphur di-
oxide boiling under
reduced pressure,
and circulated by a
second pump A. Pictet got the temperature down to — 140°, and
the pressure rose to several hundred atmospheres. On opening the
release- valve, a jet of liquid oxygen issued from it, at once boiling
away. •

Cailletet compressed the gas by a powerful pump forcing water
into a strong steel vessel, B, Fig. 86, in which the gas was contained
in a tube, T, sealed below by mercury. As water was forced into
B, the mercury was driven into the gas tube, and the gas strongly
compressed. The pressure was then suddenly released by opening
a valve which allowed the water to escape, and the gas expanded
suddenly. The expansion was so rapid (adiabatic expansion) that
the cooling produced, by the gas doing work against pressure in
expanding, reached the point of liquefaction of the oxygen. A
fog of liquid droplets was seen momentarily in the tube, at once
vanishing as heat was communicated from the walls of the latter.

FIG. 85.— Liquefaction of Oxygen by Pictet.

172

INORGANIC CHEMISTRY

Liquid air. — The liquefaction of air in bulk was effected in 1895,

independently, by Hampson in England and by Linde in Germany.

These inventors made use of a new principle, viz., the Joule-Kelvin

effect, investigated by Joule and
William Thomson (later Lord
Kelvin) in 1852-62. A compressed
gas was allowed to escape, through
a plug of silk in a boxwood tube,
into the free air, and a slight cool-
ing effect was then noticed with
most gases (air, oxygen, nitrogen,
carbon dioxide), or, with hydrogen
alone, a slight heating effect.

This temperature change is quite
different from that due to the
external work done by a gas in
adiabatic expansion. If a given
mass of gas, of volume v (Fig.
87), is forced under a pressure p1
through the plug into a space
under a lower pressure p2 (say J ^h),
it occupies a larger volume v2
(say 2 -Vj). The work done on the
gas is P^VU that done by the gas is
Pflz' If the gas obeyed Boyle's
law, PM = p2v2 (v2 = 2^ ; pl =•- 2p2),
so that no external, work would
be done on the whole, and if
no other effect were involved,
there would be no change of
temperature. Since, however, v2 is

greater than vv the molecules of the gas will have been separated,

and if an attraction exists between them, work will have been

spent on the gas in separating the molecules. With hydrogen, a

slight repulsion appears to

exist between the molecules.

The energy required for this

internal work is taken from

the heat of the gas, and a

FIG. 86. — Liquefaction of Gases by
Cailletet.

Slight cooling effect therefore Fm 87._Diagratn moating Free Expansion

results. Usually, both external of Gases.

and internal work are in-
volved. Thus, in the case of air, p2v2 is slightly larger than p{0^
since the gas is slightly more compressible than an ideal gas ;
and pl is greater than p2. A little heat is absorbed in providing
this extra work, p2v2 — p^v but more is needed for the internal

OXYGEN

173

work due to the expansion against the slight attractive forces
exerted between the molecules.

In the case of air the cooling effect is given by the formula :
Cooling effect in degrees C. = difference of pressures in atm. / 273V,

/I ^^ \ rrt i

where T^ is the absolute temperature of the air before expansion.

Thus, if air at 0°, and under a pressure of 100 atm.. is expanded

through a valve to atmospheric pressure, the fall of temperature

wiu be ? x ©' = 24-r-

Now suppose this cool air, at

- 24-7°, is allowed to sweep over the surface of a copper pipe
bringing the compressed air to the valve, by placing the latter
inside the pipe taking away the cold expanded air (Fig. 88). The
expanded air will abstract heat from the air coming
to the valve, becoming itself warmed nearly to the
atmospheric temperature. The cooled compressed air
after expansion also becomes 24-7° colder, and this
still colder air at - 49-4° sweeps over the inner
tube, reducing still further the temperature of the
compressed air coming down. The cooling effect thus
accumulates, and after an interval the air issuing
from the nozzle becomes so cold that it liquefies.
This apparatus, called a heat-interchanger, was applied
by Hampson and by Linde to the liquefaction of
air on a large scale. The two forms of apparatus
are very similar in principle. Fig. 89 shows the
apparatus used at University College, London.

Air is drawn through a purifier and filter, A, to
the compressor, BD, in which the double-acting
pistons are lubricated with water. The air is first com-
pressed in B, passes through the intercooler, C, im-
mersed in water, to the cylinder, D, where it is brought
to 200 atm. The moist air, heated by compression, passes through
the cooling coil, E, to the strong steel vessel, F, where liquid water
is deposited. Water vapour and carbon dioxide (which would
solidify and choke up the liquefier) are removed by solid caustic
soda in the vessel H, and the dry, cool, strongly compressed air
then passes down a long spiral of small-bore copper tubing in the
interchange^ K, to the expansion valve, L, which can be adjusted
from outside at M . The air, strongly cooled by expansion, sweeps
up over the outer surface of the coils in K, thereby cooling the
compressed air coming down. Liquid air finally escapes from L,
and collects in the vacuum-walled Dewar vessel, N (see below),
from the inner vessel of which it is drawn off through the valve, O.
The heat-interchanger is surrounded by a carefully lagged casing

FIG. 88.

Cooling of

Gases by Free

Expansion.

174

INORGANIC CHEMISTRY

CHAP,

to prevent inflow of heat from outside. The liquid is received in
double-walled Dewar flasks (Fig. 90), the inner surfaces of which,

FIG. 89. — Apparatus for Liquefaction of Air.

•

silvered to reflect heat, have a high vacuum between them to cut
down heat transmission to a minimum.

Liquid air, as it issues from the valve, is usually slightly turbid,
because it contains particles of solid water and carbon dioxide

from the surrounding air. If filtered
through a large filter paper it
forms a clear liquid, with a pale
blue colour. If poured out into
the air, it evaporates, producing
thick white clouds of condensed
moisture. Its temperature is about
- 190°, and when exposed to this
extreme cold many substances un-
dergo remarkable changes in pro-
perties. Lead becomes brittle, and
rubber extremely hard and brittle.
FIG." 90.-i>ewar Vacuum Vessels. Mercury is at once frozen to a

malleable solid. Raw meat, fruits,

flowers, etc., become hard, and can be reduced to powder
in a mortar. A kettle containing liquid air " boils " briskly
when placed on a slab of ice, and copious clouds of

OXYGEN

175

"steam," i.e., atmospheric moisture condensed to particles
of ice by the cold of the escaping evaporated air, are emitted
from the spout.

On standing, liquid air becomes bluer in colour ; the more
volatile colourless liquid nitrogen (b. pt. — 194°) escapes, and
sky-blue liquid oxygen (b. pt. — 182-5°) is left.

The fractionation of liquid air. — In order to obtain liquid oxygen
from liquid air, it would appear simplest to allow the liquid
to evaporate slowly in a Dewar flask, when the nitrogen would
pass off and oxygen be left. This would, however, lead to serious
loss of oxygen, as is seen from the table below :

Percentage of Percentage of Percentage of Percentage of
liquid not oxygen in liquid oxygen in gas original oxygen

evaporated.

by weight

100

23-1

37-5

50-0

60-0

67-5

77-0

88-0

evaporating.

7-5
15
23
34
42
52
70

left in liquid.
100
80
65
52
43
33
19

The gas coming from fresh liquid air contains only 7-5 per cent,
of oxygen ; when the evaporation has proceeded until the liquid
contains 50 per cent, of oxygen, or about two-thirds of the liquid
has evaporated, the gas is of the same composition as ordinary air.
It is only when 95 per cent, of the liquid has disappeared that the
gas contains 90 per cent, of oxygen, and if the remaining liquid
is evaporated to produce this rich gas, we can recover only 19 per
cent, of the oxygen originally present in the liquid air.

Linde (1902) avoided this loss by making use of a rectifying
column, in which the escaping gas is scrubbed by liquid passing
down in the opposite direction.

The air is compressed to 180 atm., cooled to — 15° in an ammonia
refrigerator, and passed to the rectifying column, K (Fig. 91),
by the tube a, which divides into three copper tubes, b. enclosed
in a wider tube, c, passing outside the column to form a heat-inter
changer. At the bottom the three tubes again unite in a single
tube, d, which passes as a spiral coil inside an iron vessel, F,
continuing as the tube / to the valve g, where the pressure
falls to about 1J atm. The cold expanded air passes to the small
reservoir, H, at the top of the rectifying column, K, whence
it leaves by the tube 1. At I' it enters the large tube c, passing
through this in the opposite direction to the compressed air coming
down the triple coil, and taking heat from the latter. The

176

INORGANIC CHEMISTRY

CHAP.

expanded air finally leaves at c', at — 15°, to the pre-cooler, escaping
from this, warmed to — 5° to — 6°, to the free air. The cooling
thus accumulates until liquid appears, which collects in H, and
finally overflows down the column K, collecting in F. This is
evaporated by the heat given out from the air inside the coil e,
and the vapour, which is mainly nitrogen, passes up the rectifying

column. Here it is
scrubbed by the de-
scending liquid, the latter
abstracting nearly all the
oxygen from the gas,
which leaves by the tube I.
The accumulating liquid
oxygen now rises in the
tube m, and gradually
evaporates, the cold gas
passing up one of the
tubes of the triple spiral,
where it takes up heat
from the incoming gas,
until it is warmed nearly
to the atmospheric tem-
perature. The gas leaves
at mf to the gas-holders,
from which it is taken
by pumps and compressed
into steel cylinders to
120 atm. When the ap-
paratus has got into
steady operation, the
working pressure is re-
duced from 180 to 90 atm.
The gas so produced con-
tains 95 per cent, of
oxygen and 5 per cent,
of nitrogen.

Claude in 1906 intro-
duced two new princi-
ples : (1) he liquefied the

air in stages, obtaining finally two liquids, one rich in oxygen
and the other in nitrogen ; (2) the expanding gas was allowed to
do work in an engine cylinder containing a piston, and the heat
equivalent of this external work was taken from it. (This had
been previously suggested by Rayleigh.) The piston is first
lubricated with petroleum ether, and then with liquid air itself.
A taller rectifying column is used, the liquid rich in nitrogen being

FIG. 91. — Linde's Oxygen Apparatus.

OXYGEN

177

discharged into the top, whilst the liquid rich in oxygen is intro-
duced at a point lower down, where the descending liquid has
become enriched to the same composition.

Claude's apparatus is shown in Fig. 92. Compressed air, cooled by
an interchanger as in the Linde process, enters A, into a vessel
containing two series of vertical pipes. The first drain into A,
and the second form a ring round the first and drain into C. Both
sets are immersed in the bath, S, which, when the machine is oper-
ating, contains nearly pure liquid oxygen.

The condensation of the compressed
air evaporates a portion of this oxygen,
part of the vapour going up the rectifying
column, D, where it is practically com-
pletely condensed, displacing nitrogen
from the liquid, and returns to S. The
rest of the vapour goes off by the pipe
G to the heat -interchanger. So far, the
process is identical with that of Linde.
The difference lies in the way in which
the compressed air is condensed. It
passes up the central group of pipes in
S, arid a liquid condenses there which is
relatively rich in oxygen, which drains
back into A. The gaseous residue passes
through the outer ring of pipes, liquefies
in them, and falls into C, the liquid
consisting almost wholly of nitrogen. The
liquid in C is then taken to the top of
the column, that in A to a lower com-
partment L, containing scrubbed liquid
of the same composition. Almost pure
nitrogen gas escapes at the top of the
column. The liquid condensed in the
inner pipes is scrubbed by air passing on.

In England there are now twelve
plants, producing an aggregate of one
million cu. ft. of oxygen gas, or 118
tons, per day, as compared with less than one-sixth of that pro-
duction in 1911. This appears large, but in Germany a single
unit plant is capable of producing nearly as much as the whole
dozen English works.

Liquid air is stored in spherical metallic vacuum vessels, holding
5-30 galls., the inner vessel being suspended by a thin metallic
neck, and the annular space highly exhausted (Fig. 93). The

FIG. 92.— Claude's Oxygen
Apparatus.

178 INORGANIC CHEMISTRY CHAP.

high vacuum is maintained by means of a tube of absorbent charcoal,
open at the end exposed to the vacuous space, and with the other
(closed) end immersed in the liquid air itself. The daily rate of
loss does not exceed 5 per cent.

When used in connection with aviation, a smaller metallic vacuum
vessel is provided with a mechanism for controlling the rate of evapora-
tion of the liquid oxygen, and a tube leading to
the inhaling mouth-piece. The controlling mechanism
consists of a siphon dropping liquid oxygen at a
controlled rate into an evaporating chamber. This
control is necessary, since great fluctuations in the
rate of evaporation are caused by movements
from higher to lower altitudes where the atmospheric
pressure is higher.

About 85 per cent, of the oxygen made is
used, in about equal proportions, for cutting

Metallic vacuum Vessel. and welding metals by the oxy-acetylene blow-
pipe (p. 189). The rest is used in medicine for

treating cases of pneumonia, gas-poisoning, etc., for oxidising

linseed oil, for maturing spirits, and in aviation.

The physical properties of oxygen.

Normal density 1-42906 gm. per Critical temperature — 118°

litre. Critical pressure 50 atm.

Boiling point — 182-9°. Specific heat of gas at 20° to 400°
Freezing point — 219° /1 2 mm. 0-2419.

Density of liquid 1-1181 at b. pt. Specific heat of liquid 0-347.

Density of solid 1 -4256 at — 222-5°. Latent heat of evaporation : 50-97

at — 182-5° ; 55-5 at — 205°

Solubility in sea-water, 78 per cent, that in pure water (p. 97).
Liquid oxygen is strongly magnetic ; the gas is weakly magnetic.

EXERCISES ON CHAPTER X

1. How may oxygen be prepared by chemical methods from air and
from water ? In what forms does the element occur in nature ?

2. What happens when the following substances are heated : mer-
curic oxide, potassium chlorate, potassium permanganate, manganese
dioxide ? The following substances are heated with concentrated
sulphuric acid : potassium dichromate, manganese dioxide, barium
peroxide. What reactions occur ? Give equations.

3. Describe the preparation of oxygen from bleaching powder, and
from potassium chlorate and manganese dioxide. How have the
reactions been explained ?

4. What classes of oxides are known ? How was the formation of
salts from acids and bases previously explained ?

x OXYGEN 179

5. What chemical methods for the manufacture of oxygen have
been proposed ? Describe the Brin process.

6. Describe briefly the principle of the method used in the liquefac-
tion of air. How is oxygen prepared from liquid air ?

7. Describe experiments which illustrate the combustion of sub-
stances in oxygen. Are the terms " combustible " and " supporter of
combustion " entirely satisfactory ?

8. It is sometimes stated that " oxygen is obtained by the evaporation
of liquid air.' Discuss this statement. What is the process actually
employed ?

9. What volumes of oxygen, measured over water at 15° and 750 mm.
pressure, would be obtained by the decomposition of : (a) 25 gm. of
potassium bromate by heat, (6) 250 c.c. of 5 per cent, hydrogen peroxide
solution by acidified potassium permanganate ?

10. One hundred c.c. of air are shaken with 1 litre of water at 2 atm.
pressure. The dissolved gas is then expelled by boiling, and the
process repeated with 500 c.c. of water. What are the volume and
composition of the gas finally obtained ?

v 2

CHAPTER XI

HYDROGEN. (H = 1)

Occurrence of hydrogen. — Although hydrogen appears to have
been discovered by Paracelsus in the sixteenth century, and an
inflammable gas (gas pingue) was described by Van Helmont
(p. 31), it was first investigated by Cavendish in 1766.

In the free state, as the gas H2, hydrogen occurs in traces in
volcanic gases ; those evolved in the eruption of Mt. Pelee in 1912
contained 22-3 per cent. H2. It also occurs in small cavities in
rock-salt, and in various minerals and rocks, such as apatite,
serpentine, gneiss, blue-clay, Peterhead granite, basalt, and beryl.
It occurs in the atmosphere, but only to the extent of about 1 part
per million, and is thus found in the earlier fractions of the gases
from liquid air. The natural gas from American oil-wells sometimes
contains up to 20 per cent, by volume of hydrogen. Meteorites,
composed chiefly of iron with nickel and cobalt, contain hydrogen
brought from the stellar space to the earth. Spectroscopic
investigation shows that the outer atmosphere of the sun consists
largely of hydrogen ; this gas is the chief constituent of the solar
prominences, which are parts of the chromosphere and are seen
during total eclipse as huge red flames of incandescent gas reaching
out from the sun's disc sometimes as far as 300,000 miles into
space. Hydrogen is produced by the decay of vegetable matter
(cellulose), owing to the activity of certain bacteria, and is present
in the gas from stagnant ponds "(p. 672).

Hydrogen occurs, however, chiefly in combination with other
elements, especially with oxygen, in the form of water, H20, which
covers such a large part of the surface of the earth. Hydrogen
is found in combination with carbon as hydrocarbons ; the gas
issuing from fissures in coal often consists of nearly pure methane,
CH4; more complicated hydrocarbons make up the petroleum,
or mineral oil, of Russia, Persia, and North America. All organic
substances in the animal and vegetable worlds, and coal, contain
hydrogen, and other hydrogen compounds found in nature are
sulphuretted hydrogen (H2S), phosphoretted hydrogen (PH3),

180

CH. xi HYDROGEN 181

ammonia (NH3), and, in volcanic gases, the halogen compounds
hydrochloric (HC1), hydrobromic (HBr), and hydriodic (HI)
acids. All acids and alkalies contain hydrogen.

Hydrogen is prepared artificially from three main sources :
(1) from water, H2O, by removal of oxygen ; (2) from acids, such
as sulphuric acid, H2SO4, and hydrochloric acid, HC1, by the action
of certain metals ; (3) from alkalies, containing the hydroxyl group,
OH, such as caustic soda, NaOH, and baryta, Ba(OH)2, by the
action of certain metals, or by electrolysis.

Hydrogen from water. — The name hydrogen was given to the
element by Lavoisier, from the Greek hudor, water. Water may
be decomposed with the liberation of hydrogen in a variety of
ways.

By electrolysis, already described on p. 56, both hydrogen and
oxygen are produced. The volumes of hydrogen and oxygen
collected should, theoretically, be in the ratio of 2 to 1 . In practice
rather less oxygen is evolved, partly on account of the greater
solubility of oxygen in water compared with hydrogen (1-8 to 1),
and partly owing to oxidation of the sulphuric acid added to the
water " to make it conduct the current."

The deficiency of oxygen is mainly due to the formation of persul-
phuric acid, H2S2O8, at the anode. Some hydrogen peroxide, H2O2,
is also formed. The presence of these oxidising agents in the liquid
round the anode may be shown by adding a solution of potassium iodide
and starch, when a blue colour, due to liberation of iodine, appears
(p. 320). The oxygen evolved also contains a little ozone, O3, and
turns blue a piece of paper dipped in potassium iodide and starch
solution. If the liquid is electrolysed hot, or phosphoric acid used
instead of sulphuric acid, no ozone is formed and the volumes are very
nearly in the ratio 2:1.

Hydrogen is also formed by the action of certain metals on water.
Sodium and potassium react violently with cold water, the latter
metal taking fire : 2Na + H20 = 2NaOH + H2.

EXPT. 70. — Press a piece of clean sodium, not larger than a small
pea, into a short length of narrow (J in.) lead tubing, closed at one
end, and hold this with tongs in a trough under an inverted jar of
water. Hydrogen collects in the jar, and the water turns red litmus
blue from the presence of caustic soda. (Explosions sometimes occur
in this experiment.) A small piece of potassium thrown on water
floats, and the hydrogen takes fire and burns with a purple flame, due
to potassium vapour. A small fused globule of caustic potash (KOH)
is left in the spheroidal condition ; this is often projected from the water
on cooling. The water contains caustic potash and turns red litmus
blue. Sodium reacts in a similar way, but the hydrogen does not take

182 INORGANIC CHEMISTRY CHAP.

fire unless the metal is kept in one place by throwing it on starch-
jelly : the hydrogen then 7 burns with a yellow flame, owing to the
presence of sodium vapour.

The action of sodium amalgam on water is much less vigorous
than that of sodium itself. The amalgam is made by dissolving
sodium in mercury ; if it contains more than 1 per cent, of sodium
it is solid at the ordinary temperature. An alloy of lead with
35 per cent, of sodium, called hydrone, may also be used.

EXPT. 71. — Small pieces of clean sodium are pressed one by one under
the surface of dry mercury in an iron mortar. Each piece dissolves
with a flash of light, poisonous fumes of mercury vapour being evolved.
The amalgam is placed in a porcelain crucible in a basin of water, under
an inverted jar of water. Gradual evolution of hydrogen occurs,
metallic mercury being left in the crucible.

H. B. Baker and L. H. Parker (1913) found that if the amalgam and
water are very pure, the action is slow, bubbles of gas appearing only
at isolated points on the surface of the amalgam. If ordinary distilled
water is added, the evolution of gas is accelerated, apparently owing to
the presence of hydrogen peroxide in the water.

Powdered magnesium, metallic calcium, and magnesium amal-
gam also decompose cold water slowly.

Boiling water is decomposed readily by magnesium powder ;
it is also decomposed by aluminium powder, and by copper-zinc
couple, prepared by the action of copper sulphate solution on
zinc : Zn -f- 2H20 = Zn(OH)2 -}- H2. The copper and zinc form
a galvanic couple, and electrolysis occurs.

EXPT. 72. — Pour a strong solution of copper sulphate over about 25
gm. of zinc dust in a 250 c.c. flask. When a deposit of copper forms,
pour off the solution and fill up the flask with previously boiled water.
Fit a cork and delivery tube and heat. Pure hydrogen is evolved.

Steam is decomposed by sodium, and by heated magnesium,
zinc, iron, cobalt, and nickel. Copper and lead do not decompose
steam.

EXPT. 73. — Insert a piece of burning magnesium ribbon into a large
flask in which water is boiling vigorously. The metal burns brightly
in the steam, and the escaping hydrogen burns at the mouth of
the flask : Mg -f- H2O = MgO + H2. Steam may also be passed over
magnesium heated in a hard glass tube : the metal burns, but the tube
usually cracks.

The decomposition of steam by red-hot iron has already been
described (p. 55). Black oxide of iron, or ferroso-ferric oxide,

xi HYDROGEN 183

J^C3^4J ig formed, but the decomposition of the steam is never
complete. A state of chemical equilibrium is set up :

3Fe + 4H20 z± Fe304 -f 4H2.

The reaction is reversible, and if hydrogen is passed over oxide
of iron, metallic iron and steam are formed. The same mixture
of hydrogen and steam results at a given temperature whether
steam is passed^over heated iron, or hydrogen over heated oxide
of iron, the mE^tioii of hydrogen decreasing with rise of tem-
perature : ^

Ratio H2/H2O by volume : 20-9 5-6 2-78 2-00
Temperature: 200° 444° 860° 918°

This process is used in the manufacture of hydrogen (p. 707). In
the Lane process reduced iron, from burnt pyrites or spathic iron
ore (p. 503), is heated in vertical iron retorts, and steam blown
through at 600-850°. The oxide of iron formed is then reduced
again with water-gas (a mixture of hydrogen and carbon monoxide,
formed by passing steam over red-hot coke, p. 707). The steaming
and reduction processes alternate. The gas contains 98 per cent,
of hydrogen, a little carbon monoxide being formed by the action
of steam on carbon deposited from carbon monoxide during
the reduction process : 2CO — C02 + C ; C + H2O =±:C» + H2.
In the Bergius process iron is heated with water in a bomb^to 300°
under 100 atm. pressure : Fe + H20 ^± FeO -f- H2. Carbon,
with a little thallium salt in the water, may be used : C + 2H20 =
C02 + 2H2.

Special processes for the manufacture of hydrogen include the actidii
of water on hydrolith, or calcium hydride :

CaH2 + 2H20 = Ca(OH)2 + 2H2.

Hydrogenite is a mixture of 25 parts of silicon, 60 parts of caustic soda,
and 20 parts of slaked lime : when ignited it burns, evolving
270-370 litres of hydrogen per kgm., and leaving sodium and calcium
silicates. In the silicol process, powdered silicon, or an alloy of silicon
with iron, is treated with a strong solution of caustic soda :

Si + 2NaOH + H2O = Na2SiO3 (sodium silicate) + 2H2.

The manufacture of hydrogen from acetylene (p. 669), and from water-
gas (p. 707), is considered later. Hydrogen is also manufactured by
the electrolysis of caustic soda solution with iron or nickel electrodes,
and as a by-product in alkali manufacture (p. 296).

Hydrogen from acids. — Acids are decomposed by many metals
with liberation of hydrogen. Nitric acid does not give hydrogen
unless it is very dilute and magnesium is used : all other metals
give various oxides of nitrogen, ammonia, etc., but no hydrogen.
The rate of liberation of hydrogen with a particular metal depends

184

INORGANIC CHEMISTRY

CHAP.

FIG. 94. — Experiment to compare Rates of Evolution of
Hydrogen by Zinc from different Acids.

largely on what is known as the strength of the acid, a property
which will be more exactly defined later. This must not be con-
fused with the concen-
tration of the acid.

EXPT. 74. — Arrange
three flasks with delivery
tubes under graduated
tubes in a pneumatic
trough (Fig. 94). In each
place 5 gm. of zinc, and
pour in 50 c.c. of normal
solutions (p. 970) of hy-
drochloric (36-2 gm. per
litre), sulphuric (49 gm.
per litre), and acetic (60
gm. per litre) acids. All
these solutions contain

1 gm. of acidic hydrogen per litre. Add 1 c.c. of dilute copper

sulpha'te to each, and after a minute has elapsed fit on the corks and

observe the rate of collection of gas. The

" strong " acids (hydrochloric and sul- | — [

phuric) react much more rapidly than the

" weak " (acetic), and hydrochloric acid

more rapidly than sulphuric.

The usual laboratory method for the
preparation of hydrogen is to act on
granulated zinc with dilute sulphuric
or hydrochloric acid :

Zn -f H2S04 =

ZnS04 (zinc sulphate) + H2.
Zn + 2HC1 =

ZnCl2 (zinc chloride) -f H2.

EXPT. 75. — A tubulated bottle is one-
third filled with granulated zinc and fitted
with a tap -funnel and delivery tube (Fig. 95 ).
Diluted hydrochloric acid (1 vol. of con-
centrated acid to 4 volumes of water), is
dropped in. If very pure zinc is used,
the reaction is slow, but it may be ac-
celerated by adding a few drops of a

solution of copper, nickel, or cobalt sulphate, or platinic chloride, when
a zinc -metal couple is formed (p. 182). The gas is collected over
water in gas jars. Before collecting the hydrogen care must be taken to

FIG. 95. — Preparation of Hydrogen.

HYDROGEN

185

allow all the air to be displaced from the apparatus : a little gas collected
in a test-tube should burn quietly with a blue flame, and not explode,
as is the case when air is still present.

The preparation may also be carried out with sulphuric acid
(1 vol. of concentrated acid -f 5 vols. of water), or iron turnings may
be used instead of zinc : Fe + H2SO4 = FeSO, + H2. The gas then
has an'unpleasant smell, due to hydrocarbons derived from iron carbide,
Fe3C, in the metal, and burns with a greenish flame. The solution in
the flask, after nitration and slight evaporation, deposits on cooling green
crystals of ferrous sulphate (" green vitriol "), FeSO4,7H2O.

If the solution of zinc in sulphuric acid is filtered from black particles
of carbon which were contained in the metal, slightly evaporated and
set aside, colourless prismatic crystals
of zinc sulphate ("white vitriol"),
ZnSO4,7H2O, separate.

Purer gas may be obtained by the
action of very dilute sulphuric or
hydrochloric acid on magnesium :
Mg + H2S04 = MgS04 + H2 ; or by
the action of a solution of mer-
curic chloride, slightly acidified with
hydrochloric acid, on aluminium:
2A1 + 6H20 = 2A1(OH)3 + 3H2. This
gas is odourless. The liquid mercury
deposited on the aluminium prevents
the formation of a protective film of
aluminium hydroxide.

Instead of a flask, a Kipp's apparatus
(Fig. 96) may be used, the metal
being placed in the central globe B
and acid poured in the top funnel until the lower bulb A is full,
and the metal covered with acid. When the tap E is closed evolution
of gas continues until the liquid is forced by pressure partly into the
upper globe, and the metal is brought out of contact with the liquid,
when the action ceases. .

The gas from commercial zinc and acid may be purified by passing
over red-hot copper turnings, or through wash-bottles containing a
saturated solution of potassium permanganate, followed by bottles
containing a 5-10 per cent, solution of silver nitrate. Impurities
such as sulphuretted, phosphoretted, and arseniuretted hydrogen,
oxides of nitrogen, sulphur dioxide, volatile hydrocarbons, and (if
red-hot copper is used) oxygen from the air, are thus removed. A
little nitrogen is left. If permanganate is used, oxygen remains, but
may be removed by a solution of chromous chloride (p. 166), or by passing

FIG. 96. — Kipp's Apparatus.

INORGANIC CHEMISTRY

CHAP.

FIG. 97.— Drying Tower.

over heated platinised asbestos. On a large scale, bleaching powder, or
a solution of bromine, is used to remove arsenic compounds from the gas.
The gas is dried by passing over granular calcium chloride, or sticks
of caustic potash, in a tower (Fig. 97) ; final drying may be effected
by phosphorus pentoxide alternating with plugs
of glass-wool in a U-tube (Fig. 98). Sulphuric
acid should not be used, as sulphur dioxide, and
sulphuretted hydrogen, are formed : H2SO4 + H2
= SO2 + 2H2O.

Hydrogen from alkalies. — A "solution of
caustic soda or potash readily dissolves zinc
or aluminium on warming, with evolution
of hydrogen :

Zn + 2KOH ==

K9Zn02 (potassium zincate) -4- H?.
2A1 + 2NaOH + 2H20 =

2NaA102 (sodium aluminate) -f- 3H2.

Hydrogen prepared in this way is much
purer than that from acids, and has no smell.

EXPT. 76. — Heat about 25 gm. of granulated
zinc with a 30 per cent, solution of caustic soda

in a flask, and collect the hydrogen. The action is more rapid if
iron filings are. added : these are unchanged, and
probably form a galvanic couple with the zinc.

Ten gm. of aluminium turnings may also be
dissolved in dilute caustic soda solution by
warming.

Pure hydrogen is evolved from the negative
electrode by the electrolysis of a warm
saturated solution of barium hydroxide in a
U-tube with platinum electrodes (Fig. 99).
This is sealed to U -tubes containing pieces
of caustic potash, followed by tubes of pure
phosphorus pentoxide, to dry the gas.

If the hydrogen from the U-tube is first
passed over heated platinised asbestos, traces of
oxygen from air -leaks are burnt to water, which is taken up in the
drying train. A little nitrogen is left, which is removed by passing the
gas into an evacuated bulb containing palladium foil, previously heated
in a vacuum. This readily absorbs more than 600 times its volume
of hydrogen, but does not absorb nitrogen or any other gas. The
residual nitrogen is pumped out of the bulb, and the latter then heated
to dull redness. Perfectly pure hydrogen is evolved.

FIG. 98. — Phosphorus
Pentoxide Drying Tube.

HYDROGEN

187

Nearly pure hydrogen may also be prepared by passing steam
over sodium, or by electrolysing dilute sulphuric acid with an
anode composed of a pool of zinc amalgam, which absorbs the
nascent oxygen liberated at the anode :

Zn + 0 = ZnO ; ZnO + H2SO4 = ZnSO4 -f H2O.

The physical properties of hydrogen. — Pure hydrogen is a colour-
less, odourless, tasteless gas. It does not support respiration, but
is not poisonous. (Impure hydrogen, containing arseniuretted
hydrogen, is poisonous.) Its molecular formula is H2 = 2.

Hydrogen is the lightest gas known, its normal density being
0-08987 gm. .per litre. It is sparingly soluble in water, and the
solubility is not greatly affected by temperature. Solubility
coefficient in water: 0°, 0-0215; 10°, 0-0198; 15°, 0-0190;
20°, 0-0184.

The spectrum of hydrogen, obtained in a Geissler tube, consists
essentially of four bright lines, although a large number of other lines

FIG. 99. — Preparation of pure Hydrogen by the Electrolysis of Barium Hydroxide solution
and absorption in metallic palladium.

are present : a red line Ha (Fraunhofer's C), 6563 A.; a blue line, Hy,
4340 A. ; a greenish-blue line, H^ (Fraunhofer's F), 4861 JL ; and an
indigo line, H5, 4102 A.

(1 Angstrom unit = JL = 10~10 metre, is the unit of wave-length,
see Chap. XXXVI.)

These lines are frequently used in calibrating spectroscopes or re-
fractometers.

Hydrogen is a good conductor of heat as compared with other
gases ; its conductivity is about five times that of air. Its specific
heat is also abnormally high : cp = 2-35 at 0°. If a spiral of
platinum wire/heated to redness by an electric current, is inserted
into an inverted jar of hydrogen, the wire ceases to glow, on account
of the increased loss of heat to the gas. According to Langmuir
(1912), the energy -loss from a wire at high temperatures in hydrogen
is greater than can be accounted for by conduction and convection:

188 INORGANIC CHEMISTRY CHAP.

he assumes that a slight dissociation of hydrogen into atoms occurs :
H2 ^ 2H, the reaction absorbing a large amount of heat (70-80
kg. cal. per gm. mol.).

Chemical properties of hydrogen. — Hydrogen is a combustible
gas, burning in air or oxygen to form water (p. 165) :

2H2 + 02 = 2H20.

Hydrogen also readily combines with fluorine and chlorine, and
less readily with bromine, iodine, sulphur, phosphorus, nitrogen,
carbon ; and with a few metals, such as lithium, sodium, and
calcium, it forms ^hydrides, such as NaH. The gas is not a supporter
of combustion : a lighted taper passed into an inverted jar of
hydrogen is extinguished.

By reason of its tendency to unite with oxygen, hydrogen acts
as a reducing agent. Thus, if hydrogen is passed over many
heated metallic oxides (copper, iron, lead), the latter are reduced
to the metallic condition, and water is produced : CuO -f- H2 ==
Cu -j- H2O (cf. EXPT. 42). Reduction is in this case the withdrawal

FIG. 100. — Oxy-Hydrogen Blowpipe.

of oxygen. Some oxides, e.g., zinc and aluminium oxides, are not
reduced by hydrogen.

Hydrogen and oxygen combine slowly at 180°, or in bright sunlight
at the ordinary temperature. Explosion occurs with moist gases at
550°, but if the gases are exceedingly pure and dry they may be heated
by an incandescent silver wire without explosion, though combination
slowly occurs (Baker, 1902) : the water produced appears to be so pure
as to exert no catalytic influence on the reaction. The mixture
2H2 + O2 ignites at 536° on adiabatic compression, some combination
occurring before the explosion itself (pre-flame period) : the mixture
3H2 + O2 ignites at 557°, and H2 + 4O2 at 507°, respectively. There
is no evidence of a minimum temperature of ignition for the mixture
H2 + O2, as was formerly supposed by Falk (Dixon and Crofts, 1914).

The oxy-hydrogen and oxy-acetylene blowpipes. — When oxygen and
hydrogen are supplied separately to a blowpipe jet consisting
(Fig. 100) of two concentric tubes, the oxygen being inside, a blue,
pointed, intensely hot flame is produced. Platinum wires readily
melt in this flame, which has a temperature of about 2800°.
(Carbon monoxide instead of hydrogen gives a flame temperature
of about 2600°.) If the oxy-hydrogen (or oxy-coal gas) flame

xi HYDROGEN 189

impinges on a small cylinder of quicklime, which is very infusible,
an intensely white light is emitted by the incandescent lime ;
this is called limelight, and is used in magic-lanterns, or for other
purposes requiring brilliant illumination. In recent years, however,
it has been largely replaced by the electric arc-light.

More recently, the oxy-acetylene blowpipe has come into use,
in which acetylene gas takes the place of hydrogen, and a much
hotter flame (3315°) is obtained. The flame is so hot that steam
is practically completely dissociated, and the reaction is

C2H2 + 02 = 2CO + H2.

The flame is therefore strongly reducing, which makes it very
suitable for welding metals.

In cutting iron or steel a third tube is used inside the other two
(Fig. 101), and when the metal is heated by the flame to a high
temperature, this inner oxygen jet is turned on. The iron itself
then burns brilliantly, emitting showers of sparks, and rapidly

FIG. 101. — Oxy-Acetylene Blowpipe.

fuses away. Since the oxygen jet is narrow, a very clean cut is
produced. Plates of steel 12 in. thick can be rapidly cut through
in this way.

The acetylene and oxygen are used in the proportions 1:5 vols. of
O2 : 1 vol. of C2H2, the acetylene being either generated from calcium
carbide and water in situ, or more conveniently used dissolved under
pressure in acetone, soaked in a porous material contained in steel
cylinders. (Compressed acetylene .gas is liable to explode spontane-
ously.) The porous material is called " kapok " and consists of the seed-
hairs found in the pods of a plant growing in India and Java.

Nascent hydrogen.

EXPT. 77. — If a little ferric chloride is added to a mixture of zinc and
sulphuric acid which is evolving hydrogen, the ferric salt is rapidly
reduced to a ferrous salt, as may be found by the appropriate tests
(p. 248) : FeCl3 + H = FeCl2 + HC1. No such change is produced
by bubbling gaseous hydrogen through the solution.

190 INORGANIC CHEMISTRY CHAP.

Zinc and sulphuric acid also reduce potassium chlorate to potassium
chloride, as may be found by the addition of silver nitrate.

It is supposed that the peculiar activity of the hydrogen in such
cases is due to the fact that it is nascent (new-born), i.e., in the act
of liberation from its compounds, and it was generally thought
that the nascent condition is due to the hydrogen being then in the
state of free atoms, which had not time to join up to form molecules
before interaction occurred. That the atomic state alone is
sufficient to confer activity on an element is improbable, because
the least active substances known (argon and its congeners) exist
in the free state in the atomic condition. Another theory is that
the hydrogen is given off under a great pressure at the instant of
generation, and this is supported by the observation that hydrogen
gas under pressure readily reduces some metallic salts (e.g., AgNO3)
in solution. It appears, however, that the nature of the chemical
action producing the hydrogen is of importance, because potassium
chlorate is not reduced by sodium amalgam, which is effective in
some other cases.

Zinc amalgam is often more effective than zinc alone, especially
if a trace of copper salt is added, and " couples " composed of
zinc with copper or iron (cf. p. 182) are frequently used for reduction
purposes.

Uses of hydrogen. — Hydrogen finds numerous uses in modern
industry. An air-hydrogen blowpipe is used for the autogeno'is
welding of lead sheets in the making of vitriol chambers (p. 505) ;
pure lead is used as a solder, being melted over the junction
by the flame (" lead burning "). The oxy-hydrogen (or oxy-
coaJgas) blowpipe is used in fusing quartz in making fused silica
apparatus, e.g., mercury lamps, or for fusing platinum. A
mixture of hydrogen and nitrogen is used in the- Haber process
for the synthetic production of ammonia (p. 543) :

N2 + 3H2 ^ 2NH3.-

An important use of hydrogen is in filling balloons and airships,
1 cu. m. of air weighs 1-29 kgm., 1 cu. m. of hydrogen weighs
0-09 kgm., so that each cu. m. of space filled with hydrogen exerts
in air a lifting force of 1 -29 — 0-09 =1-2 kgm.

The first hydrogen balloon left English soil on November 25th, 1793
The balloon was used in the American Civil War of 1861, and has since
been a recognised part of the equipment of an army. Dirigible airships,
both on land and sea, were largely used in the war of 1914-18, including
the familiar Zeppelin. The danger of fire in such cases is great, and
it is proposed to replace the hydrogen by helium (p. 604). The hydrogen
used in military balloons is usually transported in cylinders.

In recent years the importance of hydrogen has greatly increased,

HYDROGEN

191

Air A

owing to its use in the hardening of fats, e.g., in the preparation of
margarine from oils by treating the latter \vith pure hydrogen in
presence of finely-divided nickel, when the unsaturated liquid
fats take up hydrogen (p. 1005).

Diffusion of gases. — The hydrogen contained in an open inverted
jar rapidly diffuses out, and air enters ; this movement takes place
in opposition to gravity. Dobereiner in 1823 found
that hydrogen confined over water in a cracked flask
escaped into the surrounding air, the water rising in
the neck of the flask. Graham showed that as the
hydrogen escaped, air entered the flask, and since the
pressure inside is reduced, it follows that the hydrogen
diffuses out more rapidly than air diffuses in. If the
flask was covered with a bell- jar of hydrogen, no
change in the level of water occurred.

Graham devised a more convenient apparatus for
measuring the rates of diffusion of gases, consisting of a
glass tube closed at one end with a thin plug of plaster
of Paris. This tube is filled with mercury, which is then
displaced by hydrogen (Fig. 102). The mercury rises
in the tube, and the latter may be sunk in a jar of
mercury so as to keep the level constant. After a
certain time all the hydrogen diffuses out, and the
tube contains only air which has diffused inwards. No
further change of volume then occurs. If the volume
of residual air is measured, it gives the volume diffusing
in the same time as the whole of the hydrogen
originally contained in the tube. The inverse ratio of these volumes
gives the ratio of the times required for the diffusion of equal
volumes. In this way Graham found the following table, the velocity
of diffusion being the ratio of volumes diffusing in equal times : —

Gas

CH4

C02

Thus, the velocity of diffusion of a gas -is inversely proportional to the
square root of its density. This is known as Graham's law (1833).

EXAMPLE. — One hundred c.c. of hydrogen are confined in a diffusion
tube exposed to air. When change of volume ceases, what volume of
air will be left in the tube ?

FIG. 102.

Graham's

Diffusion

Apparatus.

Density

(Air = 1)

]

Velocity of
diffusion

(Air- 1).

x/Density.

0-069

3-78

3-83

0-559

1-34

0-971

1-015

1-014

1-1056

0-951

0-950

1-529

0-809

0-812

192 INORGANIC CHEMISTRY CHAP.

The volumes diffusing are inversely proportional to the densities,

vphime_o£Jiydrpgen _ VI '293
volume of air

.'. vol. of air = 100 x

- 26 4- c.c.

EXPT. 78. — The phenomenon of diffusion may be illustrated by the
apparatus shown in Fig. 103. A porous clay pot, such as is used in
batteries, is fitted by a rubber bung to a tube passing into a Woulfe's
bottle containing coloured water, as shown. Dipping into the coloured
water is a glass tube drawn out to a jet above. If a large beaker of

hydrogen is inverted over the clay pot,
hydrogen diffuses into the latter more rapidly
than air passes out, and the increase of
pressure causes the water to issue from the
jet in the form of a fountain. If the beaker
is now removed, hydrogen inside the porous
pot diffuses out into the air more rapidly
than air enters, so that the pressure is reduced.
Coloured water thus rises in the vertical
tube attached to the porous pot.

Liquid and solid hydrogen. — The first
serious attempt to liquefy hydrogen was
made by two Polish investigators,
Wroblewski and Olszewski, in 1884.
They cooled the gas to — 183°, and al-
lowed it to expand from 100 atm. pressure,
obtaining evidence of liquefaction, but
getting no liquid in bulk. The latter was
first obtained by Dewar in 1895, at the
Royal Institution in London. By com-
pressing hydrogen to 200 atm., cooling it

FIG. 103. — Experiment on
Diffusion.

to — 200°, and expanding it through a valve, he obtained a colour-
less liquid, readily boiling off. This -was liquid hydrogen. Olszewski
hi 1895 found that the .-critical temperature of hydrogen is
about — 234° (the accurate value is — 234-5°; the critical pressure
is 20 atm.), and that the slight heating effect produced by expansion
through a valve at the ordinary temperature (Joule-Kelvin effect)
changes, on cooling to — 80-5° at 113 atm., into a cooling effect.
This inversion point makes it necessary in the liquefaction of
hydrogen first to cool the gas strongly before expansion.

Liquid hydrogen is a colourless, transparent liquid, with the very
small density of 0-07105 at — 252*8° and 745-52 mm. It boils
at — 252 -7°. By rapidly evaporating the liquid under reduced

HYDROGEN

193

FIG. 104.
Preparation of
Solid Hydrogen.

pressure in a tube immersed in liquid hydrogen in a double Dewar

vessel (Fig. 104), its temperature is reduced to — 259°, when it

freezes to a colourless, transparent solid or a white, snow-like mass.

At the temperature of liquid hydrogen all other

gases except helium are frozen to solids which

at the extreme cold show practically no vapour

pressure.

If a Geissler tube is attached to a bulb containing
charcoal, and the latter dipped into liquid hydrogen,
the vacuum in the Geissler tube becomes so intense
that no electrical discharge will pass even with a
powerful coil (Fig. 105).

If a Jaarometer tube filled with air is inverted in
mercury, and the bent closed end dipped first into liquid
air and then into liquid hydrogen, the air in the tube
becomes solid, and the mercury rises to a height cor-
responding with an almost perfect vacuum.

If liquid hydrogen is poured into an ordinary test-
tube, a white coating of ice at once covers the outside. From this,
drops of liquid air are seen to fall.

Liquid hydrogen may be prepared in the modification of Travers'

apparatus devised by Nerrist (Fig. 106).
Compressed hydrogen from a cylinder or
compressor enters the apparatus through
the copper coil, A, and passes through
an extension, A', of the coil immersed in
liquid air in a large Dewar vessel. It
then passes, after cooling in this way,
through an extension of the coil, A", com-
posed of two coils in parallel inside a
small Dewar tube completely enclosed in
a brass vessel, B. At the end of this
coil is an expansion valve, F, similar
to those used in the Linde apparatus,
which is operated from outside. In the
tube A" the previously cooled gas is
liquefied by the cold expanded gas from

FIG. 105.— Experiment to show the the valve sweeping over the coil, and

liquid hydrogen collects in the inner

Dewar vessel. The cold hydrogen gas
passes out through a copper coil, <7, wound in contact with, the
coil A, and takes heat from the incoming hydrogen in the latter,
escaping into the free air, or to the compressor, only slightly below
atmospheric temperature. The liquid air boiling in the outer Dewar

very low temperature of Liquid

Hr.

194 INORGANIC CHEMISTRY CHAP.

vessel gives off cold air, which passes out through a copper coil, D,
wound between the two coils A and C, and also takes up heat from the

incoming hydrogen.
The brass vessel, B, is
in two parts, screwed
together, to permit of
the inner Dewar tube
being inserted. 300-
400 c.c. of liquid hy-
drogen are obtained
per hour, with a gas
velocity of 2-3 c.c.
per second, and the
use of about 300 c.c.
of liquid air.

The occlusion of
hydrogen by metals.

— Deville observed
that platinum and
iron become per-
meable to hydrogen
at a red heat, and
thence concluded
that " metals and
alloys have a cer-
7> ..tain porosity."
Llquid Thomas Grahlm
(1866-8) showed,
however, that the
penetration cannot
be due to the por-
osity of the metal, since hydrogen is practically the only gas which
exhibits the effect.

Graham filled a platinum bulb with hydrogen, and heated it in air.
In half an hour 97 per cent, of the hydrogen had passed out, but no
air entered, and a partial vacuum was produced inside the tube. Five
hundred c.c. of hydrogen passed per sq. m. per minute through a platinum
tube 1-1 mm. thick. Through a similar palladium tube the hydrogen
began to escape at 100° ; at a red heat 3993-2 c.c. of gas passed out per
sq. m. per minute. No other gas, except ether vapour, penetrated the
metal. Palladium in a glass tube was exposed to hydrogen at 90-97°
for three hours, and allowed to cool in the gas, for ninety minutes,
When the tube was heated by a flame, and the gas pumped off, the
metal yielded 643 times its volume of gas. Upwards of 500 vols. of

air

Liquid H2

FIG. 106. — Preparation of Liquid Hydrogen.

xi HYDROGEN 195

gas were given out when evacuation was carried out at 245°. The
surface of the palladium became roughened, and the metal was ren-
dered brittle. Iron absorbs 4 vols. of carbon monoxide, and a piece of
meteoric iron gave out 3 vols. of gas, 86 per cent, of which was hydrogen
— " the hydrogen of the stars."

Graham said that : " the whole phenomenon appears to be
consistent with the solution of liquid hydrogen in the metal . . .
It may be allowed to speak of this as the power to occlude (to shut
up) hydrogen, and the result as the occlusion of hydrogen by
platinum." In 1868 he modified his view, advancing the extra-
ordinary hypothesis that hydrogen was the vapour of an exceedingly
volatile metal, hydrogenium : " The idea forces itself on the mind
that palladium with its occluded hydrogen is simply an alloy of
this volatile metal, in which the volatility of one element is re-
strained by its union with the other, and which owes its metallic
aspect equally to both constituents." This hypothesis was exploded
when solid hydrogen was shown to be a transparent, glassy solid,
entirely devoid of metallic properties.

Palladium charged with hydrogan is a strong reducing agent :
it precipitates mercury from mercuric chloride solution, gives up
hydrogen to chlorine and iodine in the dark, and reduces ferric to
ferrous salts. Colloidal palladium takes up 2950 vols. of hydrogen.

EXPT. 79. — The occlusion of hydrogen by palladium is exhibited by
immersing two strips of palladium foil in dilute sulphuric acid, and
using them as electrodes. Oxygen is evolved from the anode, but no
gas is evolved from the cathode until the metal becomes charged with
hydrogen, when a stream of bubbles begins to come off. If the current
is switched off, gas continues to come off slowly from the cathode,
showing that the metal had become supersaturated with hydrogen.
If the current is then at once reversed, no gas comes from either elec-
trode for a time. The oxygen is combining with the occluded hydrogen
in the one electrode, and hydrogen is being occluded in the other.
After a time gas comes off from both electrodes.

Troost and Hautefeuille (1874) pumped off the gas from the
palladium at a given temperature, and measured the pressures
during its removal. The first portions of gas came off fairly readily,
but when COO vols. of hydrogen were left to 1 vol. of palladium,
the pressure became constant, and the rest of the gas came off at
this constant pressure. The phenomenon resembles the dehydra-
tion of a salt containing water of crystallisation (p. 204), and hence
these observers concluded that a definite hydride of palladium was
present. The same relations were observed at different tempera-
tures, which confirms the hypothesis.

o 2

196

INORGANIC CHEMISTRY

CHAP.

The density of palladium is 12, hence the ratio of the weights of
palladium and hydrogen in the metal which has occluded 600 vols. of
hydrogen is 12 : 600 X 0-00009 = 12 : 0*054. The atomic weight of
Pd is 106, hence the ratio of the atoms in palladium saturated with

hydrogen is -^: 0*054 = 2'03 :1, which is very near the formula Pd2H.

The author found (1919) that the logarithms of the dissociation pres-
sures found by Troost and Hautefeuille when plotted against the
reciprocals of the absolute temperatures gave very nearly a straight

line (Fig. 107). From the slope
of this line the heat of occlusion
of 1 gm. of hydrogen in palladium
was found to be 4568 gm. cal.
The value measured calorimetri -
cally by Mond, Ramsay, and
Shields (1897) was 4672 gm. cal...
in satisfactory agreement.

Roozeboom and Hoitsema
(1895) repeated the investi-
gations of the two French
chemists, but found that the
pressure curves in the dis-
sociation of the " palladium
hydride," at temperatures be-
tween 0° and 190°, consisted
of three parts (Fig. 108) : two
rapidly ascending portions,
joined by a nearly horizontal
but slowly rising middle por-
tion. At higher temperatures
the flat part becomes appreci-
ably shorter. It is less flat if
palladium black is used instead
3-43-5 Of f0ii. The dotted curves
give the results of Troost and

3-5

3-0

2-5

2-0

1-5

1-0

2-2

2-5 3-0

FIG. 107. — Dissociation Pressures of Hydrogen
in Palladium.

Hautefeuille. The shapes of

the curves were considered to speak against the existence of a
definite compound ; with certain reservations Roozeboom and
Hoitsema thought they indicated the formation of a solid solution.
The flat part, where the pressure is practically constant, indicates
that two solid solutions must be present.

Thus, since the pressure depends only on the temperature, the
degrees of freedom = 1 ; the number of components = 2, .*. number
of phases = 3, i.e., gas + 2 solids. (Cf. p. 106.)

Roozeboom and Hoitsema pointed out- that their hydrogen
contained a little nitrogen, which would explain the upward slope

HYDROGEN

197

of the curves : they did not consider their experiments sufficient
to decide the question.

Holt, Edgar, and Firth in 1913 call the occlusion of hydrogen by
palladium sorption, since they concluded that the hydrogen exists
partly as a condensed layer on the surface, and partly dissolved in
the interior of the metal. The second part is not usually homo-
geneously distributed throughout the metal.

They found that palladium exists in two forms, an active form, which
readily absorbs hydrogen, and an inactive form, which does not. In-
active palladium becomes active as a result of : (a) oxidation by
heating in air and reduction of the oxide film in hydrogen ; (6) heating
to 400° in hydrogen, followed
by cooling in the latter ; (c) Pressure
heating to 400° in vacuo ;
the hydrogen must then be
admitted as soon as cold, as
the metal so activated soon
loses its activity. In all cases,
heating is necessary for the
activation, hence the active
form of the metal is probably
an unstable variety, whilst
the stable crystalline form is
inactive.

The absorption of gas is at
first rapid, then becomes in-
creasingly slower. This sug-
gests that there is at first a
condensation of gas on the
surface ; when this becomes
saturated there is a slow dif-
fusion of hydrogen through
the mass of the metal.

The rate of diffusion of
hydrogen through palladium
0-3 mm. thick was 3288 c.c.
per sq. m. per minute at 200°,
and 5570 c.c. at 476°.

By pumping out a palladium tube saturated with hydrogen and
surrounded with the gas, the pressure inside was reduced to zero at the
ordinary temperature, whilst the pressure on the other side was 10-4 mm.
At 140°, with two pumps working equally on both sides, the outer
surface of the tube then lost 208 c.c. of gas, and the inside only 12 c.c.
The hydrogen therefore appears not to be homogeneously distributed
throughout the metal. The surface layer is easily removed by pumping ;
the gas in the interior is much more firmly held.

A. W. Porter (1918) has pointed out that different phenomena
may be confused under the name " occlusion " : (a) formation of a
chemical compound ; (6) simple solid solution, with or without

0-1 0-2 0-3 o 4- o-s Q-G atoms H to 1 atom Pd
FIG. 108. — Palladium and Hydrogen Curves.

198

INORGANIC CHEMISTRY

CHAP.

chemical combination ; (c) solid solution in contiguous phases
(Hoitsema) ; (d) surface condensation under molecular forces,
especially in pores ; (e) inclusion of bubbles of gas.

Most metals in the finely-divided condition absorb small quan-
tities of hydrogen, and metals prepared by electrolysis sometimes
contain occluded hydrogen.

Catalytic combustion. — Although oxygen and hydrogen do not
combine at the ordinary temperature, a jet of hydrogen is
inflamed if directed on a little platinum sponge, obtained by heating
ammonium chloroplatinate [(NH4)2PtCl6]. The same effect is
produced by a bundle of fine platinum wires, which become red-hot
and then kindle the hydrogen (Dobereiner, 1823). This action is
not shown by metals such as iron or copper, and in this case, there-
fore, the platinum exerts a catalytic action (p. 166).

Dobereiner's lamp consists (Fig. 109) of a small hydrogen generator,
composed of a bell -jar immersed in dilute sulphuric acid with a stop-
cock and jet at the top. A piece of
zinc hangs inside, and the gas gener-
ated displaces the acid until it is no
longer in contact with the zinc, when
action ceases. Opposite the jet is a
sponge of fine platinum wire enclosed
in a brass tube, and when the tap is
opened the stream of hydrogen ignites.
The activity of the platinum rapidly
falls off, but it may be renewed by
boiling the metal in nitric acid, when
impurities from the hydrogen, which
cause the loss of activity, are removed.

FIG. loo.— Dobereiner's Lamp. Faraday (1833) observed that

the combination of a mixture of

hydrogen and oxygen can also be brought about by a piece of
clean platinum foil — in some cases the gas explodes. There are
two theories to account for this catalytic activity of platinum in
bringing about the union of hydrogen and oxygen :

( 1 ) Faraday considered that both the gases formed a condensed film
on the metal surface — they might even be liquid. This was the result
of the action of surface-forces. Under the high pressure existing in
this film the gases entered into reaction. It is in fact known that
pressure enhances the activity of gases. Thus, Beketoff found that
hydrogen gas displaces silver and mercury from solutions of their salts
under 100 atm. pressure.

(2) De la Rive (1834), on the contrary, believed that chemical com-
pounds* unstable oxides, are formed as superficial layers on the metal.

xi HYDROGEN 199

These react with the hydrogen in a cyclic manner, the metal being
alternately oxidised and reduced : 2Pt + O2 = 2PtO ; 2PtO + 2H2 -
2Pt + 2H2O. There is some evidence for the formation of superficial
oxide films. Both theories persist to the present day, and it is probable
that both effects, the physical and the chemical, play a part in the action.

EXERCISES ON CHAPTER XI

1. Describe briefly four typical methods for preparing hydrogen.
How is the pure gas obtained ?

2. By what metals, and under what conditions, is water decomposed
with liberation of hydrogen ? Describe the commercial processes for
the preparation of hydrogen from water and iron.

3. For what purposes is hydrogen used ? Describe the construction
and use of the oxy-acetylene blowpipe.

4. What do you understand by " nascent hydrogen " ? Give two
experiments to show how nascent hydrogen differs from ordinary
hydrogen.

5. State Graham's law of diffusion, and describe an experiment to
show that hydrogen diffuses more rapidly than air. How many c.c.
of hydrogen will pass through a porous plug in the same time as 1 c.c. of
air ?

6. Describe briefly the methods used in the preparation of liquid and
solid hydrogen.

7. Give an account of the absorption of hydrogen by metals. What
theories have been advanced as to the nature of the products ?

8. Describe an experiment to illustrate the catalytic effect of platinum
in the combination of hydrogen and oxygen. What explanations of
the action have been given ?

CHAPTER XII

WATER

The physical properties of water. — Water exists in three states
of aggregation : solid (ice), liquid (water), and vapour (steam).
What is ordinarily called " steam " is not true water vapour,
which is invisible, but a mist of small droplets of liquid water.

EXPT. 80. — Boil some water in a flask fitted with a short bent tube.
A cloud of " steam " issues from the tube, but the interior of the flask,
which is filled with water vapour, is quite clear. If a Bunsen flame is
held below the delivery tube, the mist disappears. A short distance
from the tube is also seen to be clear in the first part of the experiment.
This consists of vapour which has not cooled to the condensing point.

Liquid water possesses a faint though distinct bluish-green colour >
which is seen when light is passed through a tube of water 2 m.
long, closed at the ends with pieces of plate glass. Ice also shows
the same colour in large masses, as in the crevices of glaciers or ice-
floes. The deep blue colour of certain clear lakes, however, appears
to be due to light scattered from fine particles of solid matter in
suspension (cf. p. 7).

Liquid water is only slightly compressible ; between 1 and 25
atm. an increase of pressure of 1 atm. reduces the volume by only
5 parts in 100,000. The expansion of water by heat is peculiar.
From 0° to 3-98°, the liquid contracts', beyond 3-98° it expands.
Thus, at 3-98° water is in a state of maximum density, and then
expands either on heating or on cooling. Owing to this property,
exposed water freezes only on the surface ; the water sinks as it
reaches 3-98°, and forms a heavier layer beneath the upper crust
of ice, through which heat is only very slowly transmitted.

The volume of 1 kgm. of water at 4° weighed in vacuo is defined
as the standard litre ; it occupies 1000-027 c.c. The volume of
1 kgm. of water at 15°, weighed in vacuo, is Mohr's litre ; it occupies
1000-91 c.c.

The density of ice is 0-9160 at 0° ; it therefore floats on water,
and water expands on freezing. Water pipes are burst on freezing ;

200

CH. xii WATER 201

the result is obvious when a thaw sets in. Cast-iron bottles filled
with water and closed with screw plugs are burst when immersed in
a freezing mixture.

The densities of water, referred to the weight of one-thousandth
of a standard litre at 4 ° as unity, are as follows :

DENSITIES OF WATER.

0° 0-99987 10° 0-99973 - 5° (supercooled) 0-99930

4° 1-00000 20° 0-99823 150° 0-917

8° 0-99988 100° 0-9584 250° 0-79

The amount of heat required to raise the temperature of 1 gm. of
water from 14 1° to 15|°, i.e., through 1°, is called the calorie. This
varies slightly with the temperature of the water ; thus at 0°
and 100° it is slightly greater than at 15°. The corresponding
amount for 1 kilogram of water is the kilogram calorie : 1 kgm.
cal. = 1000 cal. This heat may be generated by stirring the water,
and the number of units of work spent in generating 1 calorie is
called the mechanical equivalent of heat. Expressed in ergs this is
4-184 x 107 ergs per gm. cal. (1 erg is twice the energy possessed by
a mass of 1 gram moving with a speed of 1 cm. per second.) This
number, first determined by Joule, is denoted by J.

The number of calories required to raise the temperature of 1
gram of a substance through 1° under specified conditions is called
the specific heat of the substance. Thus, the specific heat of ice is
0-502.

When ice is converted into water a considerable absorption
of heat takes place, although the temperature remains constant
at 0°. This heat, which amounts to 79-77 cal. per gram of ice, is
called the latent heat of fusion of ice (or the latent heat of water).
Other pure substances possess characteristic latent heats. Simi-
larly, when water at its boiling point is converted into steam a large
absorption of heat occurs. For 1 gm. this amounts to 538 cal.,
and this is called the latent heat of evaporation of water (or the
latent heat of steam). In the reverse changes of solidification or
liquefaction exactly the same quantities of heat are evolved.

The vapour density of water just above the boiling point is slightly
greater than that corresponding with the formula H2O. The
presence of double molecules, dihydrol (H20)2, is sometimes
assumed to explain this. Liquid water is also assumed to consist
of dihydrol and hydrol (H2O) molecules in equilibrium : 2H2O ^
(H20)2. To explain the anomalous expansion below 3-98°, the
presence of trihydrol, (H20)3, molecules is assumed, which are formed
from hydrol by expansion. Ice would then consist largely of
trihydrol, which is also present in cold water. Although there is
evidence that liquid water is associated, or contains complex mole-

202 INORGANIC CHEMISTRY CHAP.

cules in equilibrium, the existence of these dihydrol and trihydrol
molecules is hypothetical. The case is further complicated by
the existence of four or five different varieties of ice formed from
ordinary ice under high pressures. These are all denser than water,
but different varieties lighter than water are also indicated.

The vapour pressure of water has already been considered (p. 74) :
the vapour pressures of ice are slightly less than those of super-
cooled liquid water at the same temperatures. Water is readily
cooled below 0° if kept at rest, and is then supercooled. In contact
with ice, or if violently agitated, it freezes, and the temperature
rises to 0°. Similarly, drops of water floating in oil are readily
heated much above 100° without vaporising, and are then said to
be superheated. (" Superheated steam " is merely steam which has
been raised above 100°, or the temperature of saturation, by passing
through heated tubes.)

The solvent properties of water have already been described. Some
chemists are of the opinion that dissolved salts are in some way

FIG. 110. — Ice Crystals.

" loosely combined " with the water to form unstable hydrates : e.g.,
NaCl -j- #H20 :^±NaCl,#H2O. It is supposed that, as a result of
the removal of hydrol molecules, further dissociation of dihydrol
and trihydrol occurs, leading to the changes of volume which take
place on solution. This hydrate theory of solution has led to no
unequivocal results, and although quite obviously plausible, it
is supported by no very cogent experimental evidence.

Ice crystallises in the hexagonal (six-sided) system. Beautiful
hexagonal crystals are seen (Fig. 110) when snowflakes are examined
on a cold slide under the microscope, and the crystalline form of ice
is also observed when a beam of light from a lantern is passed through
a slab of ice, which slowly melts. The bubbles in ice are composed of
air which was dissolved in the water, and is liberated on freezing.
In making clear ice, the freezing is carried out slowly, with agitation,
so that the air bubbles have an opportunity to escape. Rectangular

xn WATER 203

tanks filled with water are immersed in a large tank through which a
cold solution of calcium chloride (" brine "), which does not freeze
until — 30°, is circulated from a refrigerating machine (p. 547).

Efflorescence. — Many salts form definite solid chemical compounds
with water, called hydrates. Thus, if water is poured on white
anhydrous copper sulphate, the salt at once becomes blue, and heat
is evolved. If a hot solution of copper sulphate in water is cooled,
deep blue crystals of the hydrate, CuS04,5H20, called blue vitriol,
separate out. If these are exposed to dry air in a desiccator over
sulphuric acid, they fall to a white powder of the monohydrate,
CuS04,H20, which again becomes blue when moistened with
water.

Some crystalline hydrates lose water of crystallisation, and fall
to powder on exposure to air. This change is called efflorescence.
The loss of water as vapour on exposure to air shows that there must
be a certain pressure of water vapour over the salt, and this is con-
firmed by passing a crystal of washing soda, or Glauber's salt, above
the mercury in a barometer tube, when the mercury falls slightly.
The vapour pressure of a salt hydrate may be measured in this way.
It is found to be constant at a given temperature, and to increase
with the temperature, in the same way as the vapour pressure of a
liquid (p. 105).

In the system just described we have two components, viz., the
anhydrous salt, and water. Since the vapour pressure depends only
on the temperature, there is only one degree of freedom (p. 106) ;.
hence the Phase Rule, P -f- F = C -f 2, shows that the number of
phases is : 2 -f- 2 — 1 = 3. These phases are water vapour and two
solids. One solid is the original hydrated salt ; the second is the
anhydrous salt, if this is produced directly by loss of water, as is the
case with Glauber's salt : Na2S04,10H2O = Na2SO4 + 10H20 (vap.),
or a lower hydrate, as is the case with copper sulphate :

CuS04,5H20 = CuS04,3H20 + 2H2O.

When, at the ordinary temperature, the vapour pressure of water
above the hydrated salt is greater than the partial pressure of
moisture in the atmosphere, the salt will lose water continuously on
exposure to air, and will effloresce. If, on the other hand, the
pressure over the salt is not greatly different from that of atmospheric
moisture, the crystals of the salt will be stable on exposure to air.
Thus, blue vitriol does not effloresce on exposure to air, since the
vapour pressure over its crystals at 25° is only 7' 4 mm., whilst
the partial pressure of atmospheric moisture, which is usually about
two-thirds the saturation pressure at the given temperature, would
be 15 mm. If the vapour pressure over the hydrate is very small,
it may even absorb moisture from the air. Thus, ordinary granular
calcium chloride used for drying gases is CaCl2,2H2O. This has a

204

INORGANIC CHEMISTRY

CHAP.

very small vapour pressure, and absorbs moisture from gases,
forming CaCl2,6H2O, until only an exceedingly small amount of
moisture is left in the gas.

Vapour pressures of hydrates. — The existence of a definite vapour
pressure over a hydrated salt, as compared with the variable pressure
over a solution, when water is abstracted from the material, enables
us to distinguish between the two cases. A mechanical mixture of
liquid water with a solid may be distinguished from the two cases
just mentioned by the fact that its vapour pressure is that of pure
water, provided a solution is not formed. A hydrate containing
hygroscopic moisture, i.e., water in excess of its combined amount,
will show a vapour pressure equal to that of its saturated solution,
until all the excess of moisture has been lost ; the pressure will then
drop abruptly to that of the definite solid hydrate.

When the excess of moisture has evaporated, the definite
hydrate is left, say CuS04,5H2O, and the pressure falls abruptly
to A (Fig. 102). Dissociation of this hydrate then begins :
CuS04,5H20 =± CuSO4,3H2O + 2H2O, and the next lower hydrate,

say CuS04,3H2O,
is formed. The
system composed
of the two solid
hydrates,

CuS04,5H20
and CuS04,3H2G,
has, in accordance
with the Phase
Rule, a definite
pressure. With
continued abstrac-
tion of water, all
the higher hydr-
ate, CuSO4,5H20,
is converted into
the lower hydrate,

CuS04,3H2O.
When this occurs,
the pressure again
falls abruptly to
a lower value, re-
presented by C.

Dissociation into the lowest hydrate, CuSO4,H20, now begins :
CuS04,3H20 = CuS04,H20 + 2H20. This hydrate has a very small
vapour pressure, but gives off moisture in a desiccator over
phosphorus pentoxide, forming the anhydrous salt. Until all the
trihydrate is converted into the monohydrate, the pressure remains

S30

t°°
2

mm.

i
i

i
i
i

mm.

/
/

4- 5 mm.

Molecules H20

FlQ. 111. — Vapour Pressure Curves for Dissociation of a series
of Hydrates of Copper Sulphate at 50°.

xii WATER 205

constant. It falls sharply to a very low value, E, when the solid
is converted entirely into monohydrate, remains at this low
pressure until all the water is removed, and then falls to zero over
the anhydrous salt : CuS04,H2O ^± CuSO4 + H2O.

By analysing the solid when the sudden drops of pressure occur,
say at (7, the composition of the lower hydrates may be found.

The dotted curve AO represents the vapour pressure of a solid
solution (e.g., jelly).

Natural waters. — Water as it occurs in Nature contains various
impurities. The following division of natural waters is convenient :
(1) rain water, (2) river water, (3) spring, or deep well, water, (4) sea
water, and (5) mineral waters.

The impurities in natural water are of two kinds : (1) suspended
impurities, both mineral and organic ; (2) dissolved impurities,
both solids (mineral and organic), and gases. These are present in
amounts varying considerably with the particular source of the
water.

Rain water always contains impurities, especially if it is deposited
in the neighbourhood of towns where coal is burnt. Dissolved
atmospheric gases (oxygen, nitrogen, carbon dioxide), and sodium
chloride, derived from sea-spray carried inland by winds, are invariably
present. Nitrous and nitric acids, produced by electrical discharges
(lightning), are nearly always present in the forms of ammonium
nitrate and nitrite, and sometimes free ammonia occurs. In the
vicinity of towns, sulphuric acid, from the sulphur dioxide formed by
the combustion of coal (which contains iron pyrites, FeS2), is present.
The suspended impurities, chiefly soot from fuel smoke, are contained
in larger amounts in rain falling near towns, and the water must then
be allowed to settle before use. The free sulphuric acid may be
neutralised by adding a little lime-water, or by allowing the water to
stand over limestone. Melted snow contains similar impurities.

River water is rain water which has percolated through the
surface-soil, and taken up salts, organic matter, and suspended
matter such as clay. The dissolved matter is especially marked
when the water has passed through limestone or calcareous 'soil
(i.e., soil rich in calcium carbonate), because the carbonic acid
present in the rain, produced from atmospheric carbon dioxide :
C02 -f H20 ^ H2C03, dissolves the carbonates of calcium and
magnesium, forming soluble bicarbonates. These are unstable,
and are readily decomposed on boiling the water, with precipitation
of the insoluble carbonates and evolution of carbon dioxide :

CaC03+H2C03^±CaH2(C03)2; or MgCO3+H2CO3^MgH2(CO3)2.

206 INORGANIC CHEMISTRY CHAP.

EXPT. 81. — Pass a stream of carbon dioxide (washed free from acid
spray by passing through a wash -bottle containing water) into lime-
water. The latter at first becomes turbid, owing to the formation of
insoluble calcium carbonate : Ca(OH)2 + CO2 — CaCO3 -f H2O. On
continued passage of the gas, the precipitate redissolves, producing
calcium bicarbonate : CaH2(CO3)2, or CaO,2CO2 + H2O (i.e., a sub-
stance containing twice as much CO2, for the same weight of lime, as
the carbonate, CaO,CO2). On boiling the clear liquid, it again becomes
turbid, and calcium carbonate is precipitated. The reaction is therefore
reversible : CaCO3 -f H2O + CO2 ^± CaH2(CO3)2. If an equal volume
of lime-water is added to the clear bicarbonate solution, turbidity
is produced, and nearly insoluble calcium carbonate precipitated :
CaH2(CO3)2 + Ca(OH)2 = 2CaCO3 + 2H2O (or, omitting water :
CaO,2CO2 -f CaO = 2CaO,CO2). The filtrate is practically free from
calcium salts.

The presence of the bicarbonates of calcium and magnesium
produces what is called temporary hardness of water, i.e., such water
destroys soap without producing a lather, but is " softened " by
boiling.

Hard and soft waters. — The different varieties of hard soap
consist of the sodium salts of three organic acids, derived from fats :
Oleic acid, C17H33'C02H ; palmitic acid, C15H31-C02H ; stearic acid,
C17H35'C02H ; sodium oleate, C17H33'C02Na ; sodium palmitate,
Ci5H31'CO2Na • sodium stearate, C17H35'C02Na (" soft soap "
consists of the potassium salts of these acids).

These salts are soluble in water, but are slightly decomposed by
the latter, giving caustic soda :

C15H31-C02Na + H20=i±C15H31-C02H+NaOH.

This process of decomposition of a salt by water, with production of
free acid and base, is called hydrolysis. The reaction is reversible ; in
very dilute solutions, with a large excess of water, the hydrolysis
may be nearly complete, whilst in concentrated solutions the extent
of hydrolysis is small. In consequence, the actual percentage of caustic
soda in the solution is nearly the same for all dilutions ; it is auto-
matically regulated, and the soap does, not produce enough caustic
soda to injure the skin. The soap in addition lowers the surface tension
of water fairly considerably, so that the soapy water readily froths,
and particles of dirt tend to accumulate in the soapy liquid. The
detergent action of soap is thus an instance of separation by surface
tension effects (p. 10).

EXPT. 82. — Shake a little paraffin oil with distilled water : an
emulsion (p. 14) is formed, but this rapidly separates again into two
layers. Now add a little soap solution, and shake vigorously. A more
stable emulsion is formed. The detergent action of soap largely depends

xii WATER 207

on its property of emulsifying grease in this way ; the fine droplets can
then be washed away with water.

EXPT. 83. — Wash lampblack (fine soot) with petrol to remove grease,
and dry in a steam oven. If the fine powder is shaken with water, the
suspension settles on standing. But if soap solution is added, an inky
suspension is formed which does riot settle. The action of soap in
removing dirt depends on this action.

The calcium and magnesium salts in hard water cause a larger
waste of soap than corresponds with the production of the calcium
and magnesium salts of the fatty acids :

CaC03 (dissd.)+ 2 NaCO2'C17H35 = Ca(C02'C17H35)2 (ppd.)+ Na2C03.

About 0-17 Ib. of soap is required for 100 gallons of water containing
1 grain of CaC03 per gallon, instead of 0-075 Ib. (theoretical). The
slimy precipitate of calcium salts carries down with it some of the
soap, and renders it useless. It also adheres tenaciously to the skin
or fabric, and interferes with washing. The water does not acquire
the smooth feeling characteristic of a soft water (free from dissolved
calcium and magnesium salts), which is intensified by traces of alkali
from the excess of soap, but retains its harsh feeling until a large
excess of the soap has been added (hard water).

Ferrous carbonate also dissolves in water containing dissolved
carbon dioxide (carbonic acid), forming ferrous bicarbonate,
Fe(HC03)2. On boiling, a reddish-brown precipitate of ferric
hydroxide, Fe(OH)3, is thrown down, since the ferrous carbonate is
readily oxidised by the dissolved oxygen :

4FeC03 + 6H2O + O2 = 4Fe(OH)3 + 4CO2.

A similar ochre-like deposit is formed by oxidation of ferruginous
water in streams. If such water is used for washing, the slimy salts
formed with soap carry down brown ferric hydroxide, which adheres
to the fabric in spots, forming " iron-mould." This may be removed
by a hot solution of oxalic acid.

Temporarily hard waters deposit a crust or scale of calcium carbo-
nate when boiled in kettles or boilers, and this interferes with
the transmission of heat. It dissolves in hydrochloric acid with
effervescence.

Waters containing magnesium and calcium carbonates held in
solution by carbonic acid, when they fall in drops from the roofs of
caves, lose the carbonic acid by evaporation and deposit the insoluble
salts in the form of pendants, made up of several concentric layers,
and known as stalactites (Fig. 112). The drops falling on the floor
of the cave also deposit salts, and another concretion called a
stalagmite, growing upwards to meet the stalactite, is formed. Small
stalactites are formed under brickwork arches even in localities
where the water is soft. These are derived from the calcium

208 INORGANIC CHEMISTRY CHAP.

carbonate in the mortar, which is dissolved by the carbon dioxide
in rain water.

Temporarily hard water may be softened by the addition of exactly
the right amount of lime in the form of lime-water, or milk of lime
(Clark's process, 1841). Calcium bicarbonate is precipitated as
carbonate by adding an equivalent amount of lime :

Ca(HC03)2 + Ca(OH)2 = 2CaC03 + 2H2O.

But if magnesium bicarbonate is present, double the amount of
lime must be added, when the sparingly soluble magnesium

hydroxide is formed :

Mg(HC03)2 + 2Ca(OH)2 =
Mg(OH)2+2CaC03 + 2H20.
Magnesium carbonate is
appreciably soluble in water.
One gm. dissolves per litre,
as compared with 0-013 gm.
per litre in the case of calcium
carbonate ; the bicarbonates
are about thirty times as
soluble. The normal carbon-
ate would not be precipi-
tated, but the hydroxide is
much less soluble (0-01 gm.
per litre). The precipitates
are allowed to settle, and
the softened water is run off
for use. It may be filtered
through a bed of coke.

A different kind of hard-
ness is that due to the presence of the sulphates or chlorides
of calcium and magnesium, derived from the soil. These are
not precipitated on boiling, and cause what is called permanent
hardness. The water may at the same time possess temporary
hardness. If such waters are evaporated in boilers, gypsum
(CaS04,2H20) is deposited as a very hard, crystalline scale, which
seriously impedes the transmission of heat. This scale does not
effervesce with hydrochloric acid unless carbonates are also
present. Such waters cause waste of soap in laundry work for the
same reason as temporarily hard water. Permanently hard waters
are softened by adding a mixture of caustic soda and sodium car-
bonate (soda-ash, or else washing-soda, Na2C03,10H2O), when both
temporary and permanent hardness are removed :

CaS04 + Na2C03 - CaCO3 + Na2S04 (soluble).
Ca(HCO3)2 + 2NaOH = CaCO3 + Na2CO3 + 2H2O.

FIG. 112.— Stalactites.

xii WATER 209

Other materials used "in laundering are : ammonia, NH4'OH, which
acts similarly to caustic soda; and borax, Na2B407,10H2O, which
precipitates calcium borate, CaBO2, and also forms a little caustic
soda by hydrolysis : Na2B4O7 + 3H2O ^± 2H3BO3 + 2NaBO2 ;
NaBO2 + 2H20 ^± NaOH + H3B03.

Hardness is not known to be injurious to water for drinking
purposes (potable water)— in fact the presence of bicarbonates gives
the water a refreshing taste, and prevents its corrosive action on lead
pipes.

The hardness of a water is expressed in parts of calcium carbonate,
CaCO3, equivalent to the calcium and magnesium salts, per 100,000
parts of water, or else in grains per gallon (or parts per 70,000). It
is estimated by finding the volume of standard soap solution which
is required to produce a lather lasting five minutes with 50 c.c. of the
water. The soap solution is prepared by dissolving 10 gm. of Castile
soap, in fine shavings, in 250 c.c. of alcohol, on a water-bath. The
solution is made up to 1000 c.c. with a mixture of 4 vols. of alcohol and
1 vol. of water. The standard hard water is prepared by dissolving
0-5 gm. of pure Iceland spar (CaCO3) in hydrochloric acid, evaporating
to dryness, dissolving in distilled water, re -evaporating to remove HC1,
then dissolving to 1 litre in distilled water. 1 c.c. = 0-0005 gm. of
CaCO3. A given number of c.c. of this, made up to 50 c.c, with distilled
water to give the same soap standard as the hard water, gives the hard-
ness of the latter, in parts per 100,000. E.g., if 10 c.c. of the standard
hard water is required, made up to 50 c.c., to destroy the same amount
of soap as 50 c.c. of the given water, the latter contains 10 parts of
CaCO3 per 100,000 total hardness. If 50 c.c. of the given water are
boiled, filtered, and made up to 50 c.c., the residual hardness is the
permanent hardness. Temporary hardness — total hardness — per-
manent hardness.

River water. — River water, which has previously percolated
through soil, contains dissolved salts and suspended matter, both
mineral (clay) and organic, from vegetable matter. Water which
has flowed over peat, or peaty soil, contains dissolved organic acids
(crenic and apocrenic), which give it a yellow colour, and cause it
to corrode lead or iron pipes.

River water flowing over cultivated land contains, in addition
to the above impurities, ammonium salts, nitrites, nitrates, sodium
chloride, and organic matter of vegetable and animal origin contain-
ing nitrogen. The purity of the water depends on the nature of the
soil. Thames water, flowing over soil rich in limestone, contains
about 157 milligrams of calcium carbonate per litre. Trent water,
flowing over soil containing gypsum, contains SCO milligrams of
calcium sulphate per litre. The calcium sulphate of the Trent water
at Burton is of value in brewing. The waters of the Dee and Don,
draining the Aberdeen granite area, contain only traces of dissolved
calcium salts.

210 INORGANIC CHEMISTRY CHAP.

The dissolved oxygen of river water is of importance to fish.
One litre of river water, well aerated, contains about 50 c.c. of gas,
which is composed of 20 c.c. of nitrogen, 20 c.c. of carbon dioxide, and
10 c.c. of oxygen.

Spring, or deep well, water differs from river water only in having
undergone filtration through porous strata. In this way the sus-
pended matter may be largely removed, leaving the water clear.
The organic matter and nitrites may also have been more or less
oxidised, but the dissolved mineral impurities usually increase. Of
100 parts of rain, only 36 flow to the sea in rivers ; the rest is either
evaporated, or penetrates into the earth's crust, to reappear to some
extent in springs. This type of natural water is probably the best
for drinking purposes. Untreated distilled water is not suitable for
drinking ; if prepared for that purpose on board ship, it must be
aerated and certain salts added to make it palatable.

Sea water. — Sea water contains a large proportion of dissolved
solids, about 3-6 per cent, on the average, of which 2-6 per cent,
represents sodium chloride. It contains bromides, sulphates,
chlorides, and carbonates of magnesium, calcium, and potassium.
Traces of lithium, rubidium, caesium, and even of gold, are present.

Mineral waters. — Natural waters containing special constituents
not present (except in traces) in ordinary water are known as
mineral waters. They are of six kinds :

(1) Acidulous waters, containing dissolved carbon dioxide, together
with alkali bicarbonates, and common salt. The ca-rbon
dioxide may be liberated with effervescence when such waters
are slightly warmed, e.g., Apollinaris and Seltzer (i.e., Selters)
waters. Some acidulous waters contain sulphuric acid, probably
derived from the oxidation of sulphur dioxide or iron pyrites.

(2) Chalybeate, or ferruginous, waters, containing ferrous carbonate
held in solution by carbon dioxide as bicarbonate. On exposure
to air, such water deposits a brownish-red precipitate of ferric
hydroxide (p. 207). E.g., Pyrmont water.

(3) Hepatic waters (Latin hepar, sulphur), containing sulphuretted
hydrogen, H2S, and alkali sulphides, e.g., Na2S. These
waters smell of sulphuretted hydrogen, and on exposure to air
deposit sulphur as a white, milky turbidity : 2H2S + O2
= 2H2O -!- 2S. Harrogate water is of this type.

(4) Alkaline waters, e.g., Vichy water, contain sodium bicarbonate,
NaHCO3, and sometimes lithium bicarbonate, LiHCO3, which
are supposed to be beneficial in the treatment of gout.

(5) Bitter waters contain various dissolved salts. E.g., Marienbad
water (sodium sulphate) ; Epsom water (magnesium sulphate) ;
Friedrichshall and Hunyadi-Janos waters (sodium and magne-
sium sulphates).

xii WATER 211

(6) Siliceous water contains dissolved colloidal silica (SiO2) and
alkali silicates. Such waters are those of the geysers of Iceland,
New Zealand, and Yellowstone Park (America). They are
usually almost boiling, and deposit masses of siliceous sinter at
the mouth of the geyser.

Hot springs occur in various places, c.f/., Buxton (28°) and Bath
(47°). They often contain dissolved gas, including helium, and
traces of radium emanation (p. 1025), to which their medicinal
properties are attributed. The presence of traces of radioactive
substances may explain why artificial mineral waters, having
apparently tne same composition as the natural waters, do not
possess the same medicinal properties as the latter.

Bacteriology of water. — Numerous types of micro-organisms may
be present in natural water, mostly non -pathogenic. Germs of
typhoid, cholera, or anthrax may, however, be present, and these
diseases may be spread by polluted water. Sewage contamination,
or excremental matter, is indicated by the presence of Bacillus coli,
and since this is a comparatively robust organism, it may be assumed
that if it has been destroyed by sterilisation the other organisms are
also absent.

Water for drinking purposes is purified by filtration through
beds of gravel, and is freely exposed to air so as to take up oxygen.
It may also be sterilised by adding small quantities of chlorine, or
bleaching powder, the excess of which (giving an unpleasant
taste) may be removed by adding sodium sulphite. Three parts
excess of available chlorine per million destroy all coliform organisms
in a polluted water after half an hour's contact. The sterilisation of
water by chlorine has been largely used for military purposes. Treat-
ment with ozone has also been adopted as a method of sterilisation
(cf. p. 332). A potable water should not usually contain any Bacilli
coli in 100 c.c.

Action of water on metals. — Potable water is conveyed through
iron, lead, or zinc (galvanised iron) pipes. Certain waters on passing
through iron pipes lead to the growth of vegetation, which rapidly
corrodes the iron, causing ferric hydroxide to be deposited. In
time the pipes may be completely choked. Soft waters more than
hard are likely to attack iron. Lead is rapidly attacked by distilled,
or rain, water in the presence of air, forming lead hydroxide,
Pb(OH)2, which is appreciably soluble, or forms a colloidal solution.
Hard water has much less action on lead than soft water. The action
is due partly to dissolved oxygen, and partly to free carbonic acid.

EXPT. 84. — Two pieces of clean lead pipe are placed in two beakers
containing distilled water and tap-water, respectively, the metal being
only partly covered. Allow the beakers to estand for a few hours. The

P 2

212 INORGANIC CHEMISTRY CHAP.

distihed water rapidly becomes turbid, whilst the tap-water (if hard)
remains clear. Pour off the liquids, and add sulphuretted hydrogen
water. Compare the brown or black colorations, due to lead sulphide.
The water should not be filtered, as dissolved lead hydroxide is retained
to some extent by filter-paper.

Bicarbonates in water (temporary hardness) reduce the action on
lead ; free carbonic acid (e.g., rain water) increases the action.

Peaty waters, containing organic acids, act rapidly on lead or
zinc, unless neutralised by lime.

Pure water. — There is probably no substance more difficult to
obtain in a state of extreme purity than water. It is a close approach
to the alkahest, or universal solvent, of the alchemists, since it dis-
solves traces of practically everything with which it is brought in
contact. For chemical purposes, water is purified by distillation.
If the intermediate portion only of the distillate is collected in good
glass bottles, previously well steamed out to remove the alkaline
layer from the glass, the water is very nearly pure. A copper vessel,
with a pure tin or silver coil condenser, or a copper condenser without
brazing, is the best apparatus to use.

Still purer water is obtained by destroying the nitrogenous organic
matter, which gives off traces of ammonia on distillation, by passing
chlorine through boiling distilled water for half an hour. The chlorine
is boiled out, pure potash and potassium permanganate are added, and
the water distilled, the first half being rejected, and a quarter only of the
remainder collected. The process is repeated with this fraction. Or
Nessler solution (p. 875) may be added to the water, and the latter
distilled.

The dissociation of steam. — If electric sparks are passed through
steam (Fig. 113), it is decomposed to a slight extent into hydrogen
and oxygen : 2H2O z± 2H2 -f 02. The dissociation increases with
the temperature. The following table gives the percentage dissocia-
tion at different temperatures and pressures, i.e., the number of
molecules decomposed out of every 100 molecules of steam.

TABLE OF DISSOCIATION or STEAM.
T° abs.
1000 ...
1500 ...
2000 ...
2500 ...

Thus, at the melting point of platinum (1755°) and 760 mm.
pressure, about 6 molecules of steam in every thousand are dissoci-
ated into detonating gas. At 7*6 mm. pressure this number has
increased to 27.

10 atm.

1 atm.

0-1 atm.

0-01 atm.

1 -39X10-5

3-00 XlO-6

6-46 xlO-5

l-39xlO~4

1-03X10-2

2-21X10-2

4-76X10-2

0-103

0-273

0-588

1-26

2-70

1-98

3-98

8-16

16-6

XII

WATER

213

The dissociation of steam was discovered by Grove (1847), who heated
a platinum wire electrically in steam, passed sparks through steam, and
plunged the fused end of a platinum wire into water. In 1863, Deville
poured more than a kilogram of fused platinum into water, and found
that detonating gas was freely evolved. By passing a stream of moist
carbon dioxide through a. porcelain tube heated to 1300°, and absorbing
the gas in potash, he obtained 25 c.c. of detonating gas in two hours.

The combining volumes of hydrogen and oxygen. — The composi-
tion of water by weight has already been dealt with (p. 60), and
the approximate composition by volume also was considered. Early
experiments on this ratio are those of Cavendish (1781), who obtained
the ratio H/0 by volume 201 : 100 ; Gay-Lussac and Humboldt
(1805), who found 199-89 : 100 ; and Bunsen, whose numerous
determinations indicated an almost exact ratio of 2:1.

The accurate determination of the combining volumes was
attempted by Alexander Scott,
whose experiments, made in 1887-9
and 1893, at first yielded slightly
varying ratios, from 1994 : 100 to
200 : 100. The later experiments
showed that this variation was
due to a very thin film of grease
carried over from the lubrication of
the stopcocks into the eudiometer,
which took up a little oxygen
during the explosion, burning to
carbon dioxide and steam. By
using pure hydrogen, prepared by
passing steam over sodium, and
pure oxygen from silver oxide
(p. 159), and by lubricating the

stopcocks with syrupy phosphoric acid, the combining ratio, at
S.T.P., was found to be a little greater than 2 : 1, viz., 200-285 : 100.

Morley, by burning the gases in his apparatus, and measuring the
residual gas, found 200-269 : 100.

The most recent determination of the ratio is that of P. F. Burt
and E. C. Edgar, made at Owens College, Manchester (1915). A
short description of this research is given here as an illustration of
the refinements now possible in work carried out with gases. The
final result was 200 -288 : 100, agreeing with Scott's to within 3
parts in 200,000.

The special points of this research were : (1) very carefully purified
gases were used ; (2) the actual measurements were carried out at
0°, and under 1 atm. pressure, so that the temperature and pressure
corrections were eliminated. The hydrogen was prepared by the
electrolysis of recrystallised barium hydroxide (p. 186) ; it was

FIG. 113. — Dissociation of Steam by
Electric Sparks.

214 INORGANIC CHEMISTRY CHAP.

dried by phosphorus pentoxide, and further purified in two ways :
(i) by passing over charcoal cooled in liquid air, which readily absorbs
oxygen and nitrogen, but hydrogen only to a slight extent ; (ii) by
passing through a tube of palladium black to remove oxygen as
water, and then pumping the gas through the walls of a closed
palladium tube heated electrically. The palladium tube was welded
to a short platinum tube, and the latter sealed into a glass tube. This
was sealed inside a wider tube, and the palladium heated by a platinum
spiral wound on a quartz cylinder slipped over it. The palladium
was protected from mercury vapour from the pumps by plugs of
gold wire sponge. The palladium was charged with hydrogen at
100°, 300 c.c. of gas were then pumped off at 180°, and the metal
was recharged with hydrogen at 100°. The oxygen was prepared :
(1) by the electrolysis of baryta, liquefaction in fresh liquid air, and
fractionation ; (2) by heating pure potassium permanganate in
glass tubes, and washing the gas (a) with strong caustic potash
solution, (6) with saturated baryta solution, (c) with very strong
potash solution. The gas was then dried by sticks of potash, and
phosphorus pentoxide, liquefied, and fractionated.

The apparatus (Fig. 114) consisted of a glass 300 c.c. pipette, A,
sealed to capillary tubes at each end. The lower capillary was
expanded to a dead-space, B, of about 1 c.c. capacity, with a glass
levelling-point. The upper capillary led to a 3-way tap, (7. The
pressure of the gas in the bulb was equal to the vertical distance
between the mercury surface in B and in the upper chamber, D,
also provided with a levelling-point, and these two vessels were kept
at a constant distance apart by a stout glass rod sealed between them.
The manometer head passed to a mercury pump. The T-piece,
H, and the tap, J, formed a volume adjuster ; the capacity of the
pipette could be varied within narrow limits by withdrawing mercury
from J ; this mercury could be weighed, and its volume thus accu-
rately determined. The bulb and upper part of the apparatus were
enclosed in an ice-bath ; the lower dead-space was surrounded by a
small brine bath, M . The mercury for displacing the gas was con-
tained in O ; the air-catch, P, protected the pipette from air leaks
through the rubber. The volume of the apparatus, from C to the
level of the glass point in the dead-space, D, was determined by
weighing the contained mercury. The exit tubes from the oxygen
and hydrogen apparatus joined beyond the taps, X and F, in a
T-piece, Q, which divided again, one branch leading to the pump
through R and the other to the measuring pipette, A. The gas
was allowed to enter the pipette, displacing mercury into 0, until the
mercury surfaces in the dead-space and manometer stood at the
glass points. Since there was a vacuum above the mercury in the
manometer, the gas was measured under the pressure of this mercury
column, which was very approximately 1 atm. The tap X, or Y,

XII

WATER

215

was then closed, and the fine adjustment made by the pressure
adjuster, J, by which small amounts of gas could be added to, or
removed from, the pipette.

The gas had previously been allowed to attain the temperature of
the ice-bath, which took about three hours, and was then passed to
the explosion bulb, Z, by opening C and raising 0, mercury

To Pump

FIG. 114. — Volumetric Composition of Water : Apparatus of Burt and Edgar.

being displaced from Z through an air-trap, a, to the reservoir,
/3. Z had a capacity of about 1 litre. Two pipettes of hydrogen
with a little excess, measured by the pressure adjuster, were thus
passed into Z. A pipette of oxygen was then added in portions,
firing after each addition. The small residual volume of wet
hydrogen was sparked for a few minutes. The explosion vessel was
then cooled by a mixture of solid carbon dioxide and acetone to

216 INORGANIC CHEMISTRY CHAP.

freeze the water, the pressure reduced, and the residual gas sucked
off through a phosphorus pentoxide tube into a small pump, 8, a
spiral, 7), cooled in liquid air, being also interposed. The gas collected
in the small vessel, E, and its volume was measured as follows. The
pipette, A, was filled with hydrogen from the generator, and care-
fully levelled. The small volume of residual gas was then added
from E, and the pressure adjustment made by running a little
mercury from the adjuster. From the weight of this mercury the
volume of the residual gas was calculated.

The results of 59 experiments gave the ratio 2'00288 vols. of
hydrogen : 1 vol. of oxygen at S.T.P.

From the results of these experiments on the volumetric composition
of water, we can calculate the ratio of the hydrogen and oxygen by
weight from a knowledge of the densities of the gases. The weights of
1 litre of hydrogen and oxygen at S.T.P. are, according toMorley (p. 72),
0-089873 gm. and 1-42900 gm., respectively. There is some evidence
that Morley's value for oxygen is a little too low, by about 1 part in
28,000. Thus, Germann (1914), using carefully fractionated liquid
oxygen, found 1-42906, and the same number was found by Scheurer in
1913. Rayleigh had previously found 1 -42904. Morley's value for
hydrogen is probably the most exact we possess. Adopting Morley's
figures, the values of Burt and Edgar give, for the weight of oxygen
combining with 1 part by weight of hydrogen :

__ 1-42900
2-00288 X 0-08987~3

The composition of water. — The following table gives some of the
results of accurate investigations on the composition of water by weight
and volume, and the ratio of the densities of hydrogen and oxygen.

Experimenters.

Dumas and Boussingault (1841)
Regnault (1845) — corrected

Rayleigh (1882)

Cooke and Richards (1888)

Rayleigh (1889)

Cooke (1889)

Noyes (1890)

do.. (1907)

Morley(1895)

Scott (1893)

Thomsen (1895-6)

Reiser (1898)

Burt and Edgar (1916)

Ratio of
densities

0/H

(at S.T.P.)
15-9015
15-91
15-884

15-890

Ratio of

combining

volumes H/O

(at S.T.P.)

Atomic

weight

ot oxygen

(H = 1)

15-869
15-890

—

15-897

—

15-8799

15-9002

2-00269 : 1

15-8792

—

2-00285: 1

_~.

15-8878

—

15-869

15-8799
2-00288 : 1 —

xii WATER 217

These results offer an excellent example of the quantitative method
in chemical investigation. Starting from the assumption that the
composition of water is invariable, within the narrowest limits of experi-
mental error, the different experimenters set out to determine this
composition. The close agreement of the results confirms the original
assumption, which is a special case of a^ law of very great importance
(p. 110).

EXERCISES ON CHAPTER XII

1. Describe briefly the important physical properties of water in its
different states. How would you identify a specimen of water ?

2. Classify the various forms of natural waters. How may pure
water be prepared from these ?

3. What is meant by the hardness of water ? To what is it due, and
how may it be removed ? One hundred c.c. of water required 4-5 c.c.
n/50H2SO4 for neutralisation with methyl-orange. What weight of
quicklime must be added, in the form of lime-water, to soften 100
gallons of this water ?

4. W7hat varieties of mineral waters occur ? How would you test a
specimen of mineral w&ter to decide its character ?

5. What is the action of (a) distilled water, (b) hard water, on lead ?
How would you attempt to reduce the plumbosolvent action of water ?

6. Describe an experiment to illustrate the dissociation of steam at
high temperatures. What precautions are necessary to prevent
recombination on cooling ?

7. What are the exact combining volumes of hydrogen and oxygen ?
How have they been determined ?

8. The formula of water was formerly written HO. What is the
atomic weight of oxygen corresponding with this formula, and why is
the formula H2O now used ? What is the formula of liquid water ?

CHAPTER XIII

COMMON SALT. HYDROCHLORIC ACID. CHLORINE

Common salt. — After air and water, there is probably no material
so familiar as common salt, which is mentioned in the oldest historical
records we possess. It is an essential constituent of food, about
29 Ib. per head of population being annually consumed in this way.
About 13,000,000 tons of salt were produced in 1896 ; in 1907
there were nearly 2,000,000 tons made in Great Britain alone.
Common salt occurs abundantly, and is very widely distributed in
nature. It is contained in small quantities in all the primary

rocks. From these it has
passed by the action of
water to rivers, and thence
to the sea, where the water
re-evaporates whilst the
salt remains. Average sea-
water contains about 3 per
cent, of salt. The ex-
tensive deposits of rock-
salt, found in the earth in
many localities, appear to
have been produced by
the evaporation of former
seas and lakes.

FIG. 115.-Rock Salt Crystals. , R°ck-Salt, or halite, IS

the crystalline variety,

occurring in all the continents either as cubic crystals (Fig. 115),
which are colourless when pure but are often tinged yellow,
brown, or sometimes blue, by impurities, or else in large more
or less coloured masses, which have a cubic cleavage. Very
extensive deposits occur at Wieliczka (Poland), Cardona (Spain),
in Austria, Germany, and in England. The richest English deposits
are in the Cheshire district, at Northwich and Winsford, in the
upper Trias formation. The top bed at Northwich is 135-150 ft.
below the surface, and is 75 ft. thick. It is followed by a second

218

CH. xiii COMMON SALT. HYDROCHLORIC ACID. CHLORINE 219

bed 105 ft. thick, separated from the first by 30 ft. of hard marl.
Thinner beds occur below.

Besides rock-salt, there are brine springs, yielding a nearly
saturated solution of salt. A saturated solution contains 35-78
parts of salt per 100 of water at 15°, or about 26 per cent. The
solubility increases only very slowly with rise of temperature
(p. 99). From this brine, salt was prepared by the Romans during
their occupation of Britain, by evaporation in square lead pans
holding a few gallons. With the difference that flat iron pans hold-
ing several thousand gallons of brine are now used, the modern
process of salt manufacture in Cheshire is the same as that of the
Romans. •

The brine is tapped by bore-holes sunk through the marl ; if
no brine is found, water is poured down, becomes nearly saturated
with salt, and is pumped directly to the evaporating pans. Large
cavities are formed by the dissolving out of the salt deposits, and
serious subsidences of land often occur.

An analysis of Northwich brine is as follows : —

Sodium chloride (common salt) ... ... 25-790 per cent.

Calcium sulphate ... ... ... ... 0-450 ,, ,,

Magnesium chloride ... ... ... ... 0-093 ,, ,,

Calcium carbonate ... ... ... ... 0-018 ,, „

Calcium chloride ... ... ... ... 0-044 ,, ,,

Water 73-605 „

The more slowly the evaporation proceeds, the larger are the
crystals deposited in the pans. The different grades, according to
fineness, are : fine, or table, salt ; manu-
facturer's salt ; fishery salt, and bay salt
(usually in the form of floating " hoppers,"
or cubes with hollow faces, Fig. 116). In
some works the brine is evaporated in
vacuum pans under reduced pressure. These
are iron boilers heated by steam coils
(Fig. 117), the steam produced by evapora-
tion in one pan passing to the coils of the
next. The steam from the last pan, which
is under low pressure, is condensed by
injecting cold water into it at P, and re-
moving the extricated air along with the
water by a pump to preserve the vacuum. Each pan has a long leg
dipping into an open trough, into which the salt falls. The length
of this liquid column balances the vacuum in the pan, and thus acts
as a brine barometer.

In warm climates (e.g., in the South of France) sea- water is

FIG. 116. — " Hopper Crystals "
of Common Salt.

220

INORGANIC CHEMISTRY

CHAP.

evaporated in large flat ponds, called salt meadows, by the heat of the
sun ; the salt so made is called solar salt. The mother-liquor,
called bittern, contains the magnesium salts and bromides (p. 393)
of the sea-water. This process was formerly carried on, previous to
boiling, at Hayling Island, near Portsmouth, and at Lymington.
The industrial uses of common salt. — Besides its use in flavouring
food and assisting digestion, common salt finds a large number of
applications in industry. It is used in melting snow and ice on

roads, an effect due to
the lowering of the
freezing point of water
by the dissolved salt
(p. 103). Salt is used
in glazing common
earthenware, such as
drain pipes ; salt is
thrown into the furnace
in which the goods are
fired, and is volatilised.
The vapour forms a
fusible silicate (glaze)
with the silica of the
clay, and hydrochloric
acid is evolved :

FIG. 117.-Vacuum Evaporation Pans.

Na2Si03 + 2HC1. Un-
successful attempts
have been made to pro-
duce sodium carbonate
and hydrochloric acid
in this way on a large
scale. Very large quan-
tities of salt are used

f the alkali industry,

tor producing sodium

carbonate and caustic soda, and salt is also largely used in preserving
fish and other foods.

The history of chlorine. — By distilling common salt with concen-
trated sulphuric acid, Glauber (1648) obtained white fumes which
could be condensed in water in a receiver, forming an acid liquid,
called spirit of salt. The residue in the retort when dissolved
in water, deposited large transparent crystals known as Glauber's
salt. In 1772 Priestley found that the product of the action
of sulphuric acid on salt was a permanent gas, which could be
collected over mercury, but was very soluble in water. The solu-
tion of the gas was spirit of salt, which was then called the marine

xiii COMMON SALT. HYDROCHLORIC ACID. CHLORINE 221

acid, or muriatic acid (from Latin muria = brine). Lavoisier (1789),
in speaking of the acid, was able to say only that : " we have no
idea whatever of the nature of its radical, and only conclude, from
analogy with the other acids, that it contains oxygen as its
acidifying principle." Muriatic acid was, therefore, regarded as the
oxide of an unknown element.

In 1774 Scheele examined the action of muriatic acid on black
oxide of manganese, or manganese dioxide. He found that this
dissolved in the cold acid with the production of a dark brown
solution, which on warming gave of? a greenish-yellow gas, which had
a powerful odour of aqua regia, and bleached vegetable colours.
Scheele regarded this gas as muriatic acid deprived of its phlogiston
by the manganese, and since he considered hydrogen to be phlogiston
(p. 41), this amounts to the same thing as muriatic acid deprived
of hydrogen : Muriatic acid — H. This is correct.

In 1785 Berthollet found that when a solution of the new gas in
water was exposed to light, it gave off bubbles of oxygen and left
a solution of muriatic acid. In accordance with Lavoisier's theory
of acids, he therefore considered that the gas was a compound of
muriatic acid and oxygen, or oxymuriatic acid. He recognised,
however, that it was not an acid, which was a serious difficulty from
the point of view of this theory.

Gay-Lussac and Thenard in 1809 heated sodium in muriatic acid
gas, and found that hydrogen was evolved and common salt rema'ned.
The hydrogen, they supposed, came from water existing in combina-
tion in the gas, but they were unable to obtain oxygen from the latter
or to oxidise charcoal heated to whiteness in the gas. Nevertheless,
they decided in favour of Lavoisier's view, and rejected the alterna-
tive that the gas was a compound of " oxymuriatic acid," which
was really an element, and hydrogen.

The elementary nature of oxymuriatic acid was, however, strongly
urged by Davy in 1810. He heated charcoal, sulphur, and metals in
the gas, but never obtained any known oxygen compound. He
proposed to regard it as an element, and called it chlorine (Greek
chloros = pale green). In Berthollet 's experiment, the oxygen
came from the water, the hydrogen of which united with the chlorine
to form muriatic, or hydrochloric, acid : H20 -f- C12 = 2HC1 + 0.
Dry chlorine, Davy found, did not bleach. " I merely state what I
have seen," says Davy, " and what I have found. There may be
oxygen in oxymuriatic gas, but I can find none." After a little
controversy, this view was accepted.

The preparation of chlorine. — Chlorine is prepared in the labora-
tory by the .oxidation of hydrochloric acid : 2HC1 + O = H2O + C12.
The operation may be carried out in several ways, according to the
oxidising agent employed. Atmospheric oxygen, in the presence of

222

INORGANIC CHEMISTRY

a copper salt acting as a catalyst, may be used ; or substances rich
in oxygen which readily part with that element, such as manganese
dioxide, MnO2 ; potassium permanganate, KMn04 ; potassium
dichromate, K2Cr207 ; and bleaching powder, CaOCl2 : —

4HC1 + O2 = 2H20 -f- 2C12 (atmospheric oxygen) ;

4HC1 + Mn02 = MnCl2 + 2H2O + C12 (manganese dioxide) ;

2KMn04 + 16HC1 = 2KC1 + 2MnCl2 + 8H2O + 5C12 (potassium
permanganate) ;

K2Cr2O7 + 14HC1 = 2KC1 + 2CrCl3 + 7H20 + 3C12 (potassium
dichromate) ;

CaOCl2 + 2HC1 = CaCl2 + H2O + C12 (bleaching powder).

FIG. 118. — Oxidation of Hydrochloric Acid Gas by Atmospheric Oxygen with formation

of Chlorine.

EXPT. 85. — A stream of air is passed through concentrated hydro-
chloric acid in a Wouife's bottle, and concentrated sulphuric acid allowed
to drop slowly into the latter. The mixture of air and hydrochloric acid
gas is passed through a hard glass tube packed with pieces of pumice
which have been soaked in a solution of copper sulphate and dried, and
the tube is heated in a furnace (Fig. 118) to a dull red heat. The gas is
then passed through litmus solution, which is rapidly bleached by the
chlorine evolved.

EXPT. 86. — Place powdered manganese dioxide in one bulb of a hard
glass tube, leaving the other bulb empty. Pass a current of hydro-

xin COMMON SALT. HYDROCHLORIC ACID. CHLORINE 223

chloric acid gas, obtained by dropping concentrated hydrochloric acid
into concentrated sulphuric acid in the apparatus of Fig. 81, over tho
dioxide, and allow the
gas to pass into a bottle
containing litmus solu-
tion and a piece of moist
red flannel (Fig. 119).

is turned

heat the
dioxide.

The litmus

reel. Now

manganese

Moisture collects in the

second bulb, and the

bottle becomes filled

with a greenish-yellow

Wire

^ Piece of
Flannel

H-

^_~_ ~

FIG. 119.— Oxidation of Hydrochloric Acid Gas by heated
Manganese Dioxide.

gas, which bleaches the litmus and the red flannel. This is chlorine.

The usual method of preparing chlorine is to decompose hydro-
chloric acid with manganese dioxide : the mineral form, called
pyrolusite, in small pieces, is most convenient.

EXPT. 87. — One-third fill a litre flask with small pieces of pyrolusite,
and fit the flask with a good black rubber stopper, carrying a dropping

funnel and de-
livery tube con-
nected with a
wash-bottle con-
taining a little
water, to remove
hydrochloric acid
gas, and a second
bottle containing
concentrated sul-
phuric acid (Fig.
120). Pour 200 c.c.
of concentrated
hydrochloric acid
into the flask.
Notice the form-
ation of a dark
brown solution.
Heat gently on
wire-gauze, or in

a water-bath, and collect the chlorine in jars by downward displace-
ment (it is 2^ times as heavy as air). The preparation is carried out
in a good draught cupboard, as the gas has a powerful corrosive action

FTG. 120. — Preparation of Chlorine from Aqueous Hydrochloric
Acid and Manganese Dioxide.

HAP.

ited

224 INORGANIC CHEMISTRY CHAP.

on the mucous membrane. The inhalation of alcohol vapour, and diluted
ammonia gas, relieves the bad effects produced by breathing chlorine.

The action of manganese dioxide on hydrochloric acid proceeds in
two stages. The dark brown solution formed in the cold contains a
higher chloride of manganese, MnCl4 or MnCl3, which breaks up on
warming, with liberation of chlorine :

(1) Mn02 + 4HC1 = MnCl4 + 2H2O. (2) MnCl4 = MnCl2 + C12 ; or
2Mn02 + 8HC1 == 2MnCl3 + C12 + 4H2O. (2a) 2MnCl3 =
2MnCl2 + C12.

If the dark brown solution is poured into water, hydrated man-
ganese dioxide is precipitated : MnCl4 + 2H20 = Mn02 + 4HC1.

A mixture of 5 parts of powdered MnO2, 11 parts of common salt, and
14 parts of 50 per cent. H2SO4 may also be heated in a flask to produce
chlorine ; this was the method used by Berthollet (1785), but is less
convenient than that of Scheele : 4 NaCl + MnO2 + 3H28O4 = C12 +
2NaHSO4 + Na2SO4 + MnCl2 + 2H2O.

EXPT. 88. — If red crystals of potassium dichromate are heated in
a flask with concentrated hydrochloric acid, a green solution of chromic
chloride, CrCl3, is formed^ and practically pure chlorine is evolved.

EXPT. 89. — The most convenient method of preparing small quanti-
ties of chlorine is to drop concentrated hydrochloric acid slowly on crys-
tals of potassium permanganate in a flask (Fig. 81). The gas is evolved
in the cold, and may be washed with water and concentrated sulphuric
acid. When the evolution of gas ceases, a further supply is obtained on
warming the dark brown solution formed ; this becomes nearly colourless.

EXPT. 90. — If concentrated hydrochloric acid is dropped on bleaching
powder in the above apparatus, chlorine is evolved. The bleaching
powder may first be mixed with one-fourth its weight of plaster of Paris,
moistened slightly, pressed and cut into cubes, which are dried at the
ordinary temperature. These evolve chlorine if treated in a Kipp's
apparatus with concentrated hydrochloric acid, and the evolution of
gas may be controlled in the usual way (p. 185).

Pure chlorine may be obtained by heating platinic chloride,
PtCl4, or auric chloride, AuCl3 (gold chloride), in a hard glass tube.
Lower chlorides are first produced, which decompose, giving the
metals, at higher temperatures :

300° 500°

PtCl4 (platinic chloride) = PtCl2 (platiaous chloride) -f C12 = Pt + 2C12.

175° 185°

2AuCl3 (auric chloride) = 2AuCl (aurous chloride) -f- 2C12 = 2Au + 3C12.

xin COMMON SALT. HYDROCHLORIC ACID. CHLORINE 225

Cupric chloride, CuCl2, when heated to about 350°, decomposes
into cuprous chloride, CuCl, and chlorine : 2CuCl2 = 2CuCl -f- C12.
The cuprous chloride is stable at high temperatures and does not
further decompose. The catalytic action of cupric chloride in the
oxidation of hydrogen chloride by atmospheric oxygen (p. 222), or
pure oxygen gas, has been explained as follows. The cupric chloride
first decomposes, with evolution of chlorine, and leaves cuprous
chloride : 2CuCl2 = CufiCl2 + C12. By the action of hydrogen
chloride and oxygen on the cuprous chloride, cupric chloride and
water are formed : 2Cu2Cl2 + 4HC1 + 02 = 4CuCl2 + 2H20. The
cupric chloride again decomposes, and thus undergoes a cycle of
chemical changes (cf. p. 167). It may be assumed, following
Mercer, that the attraction of cuprous chloride for chlorine, with the
attraction of oxygen for hydrogen, can together decompose the
hydrogen chloride, but each acting separately is unable to effect
any change :

TT ; PI

> Cu2Cl2 + C12

gp Cu2Cl2 -> 2CuCl2 -> Cu2Cl2 + C12

Silver chloride, AgCl, on heating does not decompose, but melts at
460° to a dark yellow liquid. This conducts an electric current,
and if the electrolysis is carried on in a Jena glass U-tube with gas-
carbon poles, pure chlorine is evolved at the positive pole and silver
deposited at the negative.

The properties of chlorine. — Chlorine is a greenish-yellow gas,
the normal density of which is 3-220 gm. per litre. The relative
density at S.T.P. is therefore 35'80. The relative density cal-
culated from the atomic weight is 35-2, and the somewhat higher
observed density may indicate a slight polymerisation : 2C12 ^ C14.
The density decreases slightly with rise of temperature, and becomes
normal at about 240°, remaining at this value up to 1200° :
Temperature0 ____ 0 40 80 120 160 200 240 1200
Relative density .. 35-8 35-7 35-6 35-5 35-4 35-3 35-2 35-2.

The density at 1150° was found by Reinganum (1905) by com-
paring the volumes of gas displaced from a small quartz Victor
Meyer apparatus, in one case filled with oxygen and in the other
with chlorine. They were equal, hence no dissociation had occurred.
Meier and Crafts (1881) obtained the same result by displacing
oxygen by chlorine, or chlorine by oxygen, in a porcelain apparatus
at 1350°. Victor Meyer and Langer (1885), by burning gas-carbon
in a powerful blast of oxygen, claimed to have attained a tempera-
ture of 1700°, at which the density of chlorine fell to 29-03, which
would correspond with a 35 per cent, dissociation into atoms :
C12 ^± 2C1. Pier (1908) found a dissociation of C12 above 1450°.

226

INORGANIC CHEMISTRY

CHAP.

Chlorine, when cooled in solid carbon dioxide and ether, condenses
to an orange-yellow liquid, boiling at — 33-6°. On cooling in liquid
air, this forms a pale yellow solid, melting at — 102°. The critical
temperature of chlorine is 146°; its critical pressure is 93-5 atm.
The gas is liquefied at 0° by a pressure of 3-66 atm. ; at 20°, 6-62 atm.
pressure is required.

The chemical properties of chlorine. — The chemical properties of
chlorine may be summed up in the statement that it is a very active
element ; it combines readily with hydrogen, and directly with most
metals, and non-metallic elements except nitrogen, oxygen, and

carbon. Combination often
occurs when the elements
are brought together at the
ordinary temperature, often
with the production of
flame, or incandescence.

The reaction with metals,
which occurs violently with
moist gas, does not take
place if the chlorine is dry,
except in the case of mer-
cury, which completely ab-
sorbs pure dry chlorine.
Sodium may be distilled
in dry chlorine without
reaction taking place
(Wanklyn, 1883). The
reason for the action of
moisture is not known. In
the following experiments,
therefore, unless otherwise directed, moist chlorine is to be used.

EXPT. 91. — Sprinkle a little finely powdered arsenic, and antimony,
into jars of chlorine. The substances burn brilliantly, producing
poisonous fumes of the chlorides AsCl3, SbCl3, and SbCl6.

EXPT. 92. — A piece of phosphorus in a deflagrating spoon ignites
spontaneously in chlorine, burning with a pale flame, and producing
fumes of the chlorides PC13 and PC16.

EXPT. 93. — Pass chlorine over a piece of sodium heated in a hard
glass bulb tube (Fig. 121). When strongly heated, the metal catches
fire and burns with an exceedingly brilliant yellow flame, producing
white sodium chloride, NaCl.

FIG. 121. — Combustion of Sodium in Chlorine.

xin COMMON SALT. HYDROCHLORIC ACID. CHLORINE

227

EXPT. 94. — A carefully dried bolt-head flask fitted with a rubber
stopper and stopcock is loosely filled with leaves of Dutch metal (an
alloy of composition copper 80 + zinc 20). The flask is then evacuated
and filled with dry chlorine from a bell -jar standing over concentrated
sulphuric acid (Fig. 122). No action occurs. The stopper is removed
and a drop of water is allowed to fall into the flask : the metal at once
catches fire and burns, producing yellow fumes (CuCl2 and ZnCl2). A
spiral of German-silver wire, tipped with Dutch metal, ignites and burns
when introduced into a jar of moist chlorine, throwing off a shower of
sparks.

EXPT. 95. — A jet of hydrogen burning in air continues to burn, with
an enlarged greenish flame, when introduced into a jar of chlorine

• FIG. 122.— Filling a
flask containing Dutch
Metal with Dry
Chlorine.

FIG. 123. — Combustion of Hydrogen in Chlorine.

(Fig. 123), producing fumes of hydrochloric acid : H2 -f C12 = 2HC1.
These redden moist litmus paper. A jet of chlorine burns when intro-
duced into an inverted jar of hydrogen which is burning at the mouth.

EXPT. 96. — A piece of dry red flannel, and some dry litmus paper,
suspended in a jar of chlorine, into which some concentrated sulphuric
acid has been poured, are not bleached. If a little steam is passed in,
bleaching at once occurs (cf. p. 223).

EXPT. 97. — A burning taper plunged into a jar of chlorine burns with
a small dull-red flame, clouds of black carbon and white fumes of hydro-
chloric acid being evolved. Wax is a mixture of hydrocarbons,

Q 2

228 INORGANIC CHEMISTRY CHAP.

CnH2n + 2; the chlorine removes the hydrogen, forming HC1, and sets
free the carbon, with which it does not combine directly. Char-
coal heated to redness in a deflagrating spoon ceases to burn in
chlorine.

EXPT. 98. — A mixture of 2 vols. of chlorine and 1 vol. of ethylene
(p. 675), C2H4, when ignited, burns with a red flame, emitting dense
black clouds of carbon : C2H4 + 2C12 = 20 + 4HC1.

EXPT. 99. — A mixture of 2 vols. of chlorine and 1 vol. of methane
(p. 672), CH4, prepared out of direct sunlight, ignited with a taper,
burns with a feeble whistling noise, giving fumes of hydrochloric acid
and a cloud of carbon : CH4 + 2C12 = C -f 4HCL

EXPT. 100. — A little turpentine, C10H16, warmed in a test-tube and
poured on filter-paper, catches fire when plunged into chlorine, giving
a black cloud of carbon and fumes of hydrochloric acid.

Chlorine combines with the gases sulphur dioxide, S02, carbon
monoxide, CO, and ethylene, C2H4, producing sulphuryl chloride,
SO2C12, carbonyl chloride (phosgene), COC12, and ethylene dichloride,
C2H4C12, respectively. The carbon monoxide and sulphur dioxide
react with chlorine in presence of animal charcoal ; ethylene com-
bines directly with chlorine if the mixture of gases is exposed to
light, an oily liquid being formed.

Chlorine water. — Chlorine is fairly soluble in water, 2 volumes
of the gas dissolving in 1 volume of water at 15°. The solution,
which may be prepared by passing chlorine through cold water in
Woulfe's bottles, is pale yellow in colour, and smells strongly of the
gas. It is called chlorine water. The solution possesses bleaching
and oxidising properties. It precipitates sulphur from a solution of
sulphuretted hydrogen : H2S + C12 = 2HC1 + S ; it liberates
iodine from a solution of potassium iodide : 2KI -f- C12 = 2KC1 -\- I2,
but with an excess of chlorine water the iodine dissolves, forming
iodine chloride, IC1. A solution of sulphur dioxide (sulphurous acid)
is oxidised to sulphuric acid : 2H20 + C12 + S02 = H2S04 + 2HC1.

When a flask of chlorine water, inverted in a basin of the same
liquid, is exposed to bright sunlight, it is decomposed with evolution
of bubbles of oxygen, and a solution of hydrochloric acid is left :
2H20 -f 2C12 = 4HC1 + 02.

Chlorine hydrate. — If chlorine is passed into water cooled in ice,
greenish-yellow crystals separate. This substance, discovered by
Berthollet in 1785, is chlorine hydrate : its composition has been
variously stated to be C12,10H20 (Faraday, 1823), C12,8H20 (Rooze-
boom, 1884), and C12,7H2O (de Forcrand, 1902). When gently
warmed, the crystals melt with effervescence, and chlorine is evolved.
If the experiment is performed in the dark, the gas, after drying,
is perfectly pure (Harker, 1892).

xin COMMON SALT. HYDROCHLORIC ACID. CHLORINE 229

EXPT. 101. — If crystals of chlorine hydrate are sealed up in one
limb of a strong bent tube, and the other limb is cooled in ice and salt
(Fig. 124), liquid chlorine distils into the cooled part of the tube when the
other is warmed to about 30°.

Hydrogen chloride, or hydrochloric acid, HC1. — Chlorine and
hydrogen form only one compound, hydrogen chloride, or 'hydro-
chloric acid, HC1. This is formed by the combustion of hydrogen
in chlorine, but is usually prepared by the action of slightly diluted
sulphuric acid on common salt : NaCl -f- H2S04 = NaHS04 + HC1.
One only of the two hydrogen atoms of sulphuric acid is expelled,
and the acid salt, NaHSO4, sodium hydrogen sulphate, or sodium
bisulphate (Na20,2S03,H20) is formed unless the temperature is
higher than can conveniently be attained in a glass flask. This salt,
which contains one of the hydrogen atoms of the sulphuric acid, has
a strongly acid reaction in solution, and neutralises caustic soda, or
sodium carbonate, with formation of the normal salt, Na2S04. This
crystallises from water
as Na2S04,10H2O, which
is Glauber's salt. If the
acid salt is strongly
heated with common
salt, the remaining hy-
drogen atom is displaced
as hydrochloric acid, and
the normal salt formed :
NaHS04 + NaCl =

Na2S04 + HC1.
The hydrogen of sul-
phuric acid can be dis-
placed in two stages, with formation of acid salts and normal
salts, hence sulphuric acid is called a dibasic acid. Hydrochloric
acid, which contains only one atom of hydrogen, forms only one
series of salts, the normal salts, and is called a monobasic acid.

EXPT. 102. — The preparation of the gaseous acid is carried out in the
apparatus shown in Fig. 125. Common salt is placed in the flask, and
covered with diluted sulphuric acid, prepared by adding 11 vols. of
concentrated sulphuric acid to 8 vols. of water, and cooling. When the
flask is gently heated on wire gauze, a steady stream of hydrochloric
acid gas is evolved. This is passed through a small wash-bottle con-
taining concentrated sulphuric acid, and then collected in dry jars by
downward displacement, since it is 1-27 times as heavy as air, and is
very soluble in water. It may also be collected over mercury. When the
jar is full of gas, dense white fumes issue from the mouth. These are
formed from the gas and atmospheric moisture, producing minute drop-

. 124.— Liquefaction of Chlorine.

230 INORGANIC CHEMISTRY CHAP.

lets of solution, which have a lower vapour pressure than the partial
pressure of water vapour in the air. The dry gas is quite transparent.

If the gas is passed into a flask of distilled water, kept cool by running
water over the outside from a ring of perforated lead pipe placed over
the neck (Fig. 126), an aqueous solution of the acid — spirit of salt— is
produced. Each bubble of gas at once condenses as it leaves the
delivery tube, and a considerable amount of heat is given out. The
concentrated solution fumes strongly in the air.

Hydrogen chloride is very soluble in water. When 1 kgm. of
water is saturated with the gas at 15° it increases in weight to 1-75
kgm., and the density is 1-22. It contains about 43 per cent, of

FIG. 125. — Preparation of Hydrogen Chloride.

FIG. 126. — Preparation of a Solution
of Hydrogen Chloride.

HC1 ; the commercial acid contains about 40 per cent., its density
being 1-20.

Densities of aqueous solutions of hydrochloric acid at 15°.

Per cent. HC1. Density. Per cent. HC1.

10 1-1490 29-35

15-84 1-1696 33-39

20-29 1-1901 37-23

Density.
1-0491
1-0784
1-1014
1-1271

25-18

1-2002

39-15

The most convenient method of obtaining the gas is to drop con-
centrated hydrochloric acid into concentrated sulphuric acid by
means of a tap-funnel. A rapid stream of gas is evolved.

EXPT. 103. — The great solubility of hydrochloric acid gas in water
may be demonstrated by the fountain experiment. A large round-
bottomed flask is filled with the gas and fitted with a rubber stopper

xrii COMMON SALT. HYDROCHLORIC ACID. CHLORINE 231

carrying a tube drawn out inside the flask into a jet. The flask is
inverted and connected with a tube dipping into water coloured with
blue litmus contained in a second large flask, as shown in Fig. 127. By
blowing into the short tube on the second flask a drop of water is forced
into the upper flask. The gas is instantly dis-
solved, and a vacuum is formed. The water in
the lower flask is therefore driven in the form of a
fountain into the upper flask, and the litmus is
turned red by the acid solution formed.

EXPT. 104. — Hydrochloric acid collected in
jars will be found to extinguish a taper, and
to be non-inflammable. Burning sulphur and
phosphorus are extinguished in the gas, but a
little potassium burning in a deflagrating spoon
continues to burn in the gas. If the potassium
is heated in a hard glass tube in a current of
the gas, it burns, forming potassium chloride,
and the hydrogen evolved may be ignited :
2HC1 + 2K = 2KC1 + H2.

The composition of hydrochloric acid. — It is

easily shown by experiment that hydrochloric
acid gas contains half its volume of hydrogen.

EXPT. 105. — Col-
lect the gas in a care-
fully dried tube over
dry mercury (Fig.
128). By means of
a bent pipette intro-
duce a drop of water

into the tube. The gas at once dissolves,
and the mercury rises and fills the tube.
Now pass a piece of magnesium ribbon into
the tube. It rises through the mercury, and
on contact with the aqueous acid dissolves,
with liberation of hydrogen. This fills half the
tube. If the latter is closed with the thumb,
and inverted, the gas may be ignited with
a taper : 2HC1 + Mg = MgC32 + H2.

EXPT. 106. — Electrolyse concentrated hydrochloric acid, saturated
with common salt, in the apparatus shown in Fig. 129, using electrodes of
gas-carbon, since chlorine attacks platinum. The chlorine evolved at
the anode at first dissolves in the liquid, but when the latter becomes
saturated, equal volumes of hydrogen and chlorine are evolved. These

FIG. 127. — Demonstration
of the Solubility of
Hydrogen Chloride.

FIG. 1 28. — Decomposition

of Hydrogen Chloride by

Magnesium.

232

INORGANIC CHEMISTRY

CHAP.

may be recognised by the inflammability of the former, and the action
of the latter on a piece of moist litmus paper, which is bleached.

EXPT. 107. — Fill the closed limb of the U-tube shown in Fig. 130 with
dry hydrogen chloride to the lower stopcock,
by admitting the gas through the upper stop-
cock, and running out the dry mercury from the
tube. Close the lower stopcock, pour out the
mercury, and replace it with liquid sodium
amalgam. Open the stopcock, agitate the gas
with the amalgam, and allow the apparatus to
stand. A white crust of sodium chloride is
formed, and the volume of the gas, after
levelling, is found to be diminished to one-half.
Pour mercury into the open limb of the U-tube,
and displace the gas through the stopcock.
It will be found to be inflammable, and is
hydrogen.

EXPT. 108. — Fill one half of a strong glass
tube, provided with three stopcocks, as shown
in Fig. 131, with chlorine by passing the gas
through whilst the middle, three-way, stopcock
is open to the air. Fill the other half with

hydrogen in thje same way. Take the tube into a room with
diffused daylight, open the middle stopcock, and allow the gases to mix.
After exposure to diffused daylight for a few hours,
the greenish-yellow colour of the chlorine disap-
pears. If one of the end stopcocks is opened under
mercury, no gas escapes and no mercury enters,
hence the volume is unchanged. If the tube is
opened under water, the latter enters and fills
the tube. The liquid is acid, and contains hydro-
chloric acid. This experiment shows that 1 vol. of
hydrogen -f- 1 vol. of chlorine = 2 vols. of hydrogen
chloride.

EXPT. 109. — Pass the mixture of hydrogen and
chlorine evolved by the electrolysis of concentrated
hydrochloric acid, saturated with common salt
/'Fig. 134), through a glass tube fitted with two stop-
cocks and platinum firing wires (Fig. 132), The
electrolysis should be allowed to proceed for about
half an hour before collecting the gas, so as to
saturate the liquid with chlorine, and the tube filled in a dark room
with a photographic ruby lamp. Support the tube in a clamp behind

FIG. 129.— Electrolysis of
Hydrochloric Acid.

FIG. 130. — Decom-
position of Hydrogen
Chloride by Sodium
Amalgam.

xm COMMON SALT. HYDROCHLORIC ACID. CHLORINE 233

a strong glass screen, and explode the gas by a spark from a coil.
When the tube is cool, open one stopcock under mercury. No gas
bubbles out, and no mercury is drawn in, hence the volume is un-
changed by combination. Pour a layer of previously boiled water

FIG. 131.— Tube for Combination of FIG. 132.— Explosion Tube for

Hydrogen and Chlorine. Hydrogen and Chlorine.

over the mercury and raise the tube so that the open stopcock dips
into the water. The gas dissolves, and the tube is filled with water.

These experiments prove that 1 volume of hydrogen combines
with 1 volume of chlorine to produce 2 volumes of hydrogen chloride.
The (corrected) relative density of hydrogen chloride is 18-1, hence its
molecular weight is 36-2. Thus, 22-2 litres at S.T.P. weigh 36-2 gm.
This volume contains 11-1 litres, or 1 gm. of hydrogen, and therefore
36-2 — 1 = 35-2 gm. of chlorine. The formula is HCl^. But in
all the volatile compounds of chlorine, never less than 35-2 parts of
chlorine are contained in a molecular weight, hence 35 -2 is the atomic
weight of chlorine, and the formula of hydrogen chloride is HC1.
From the density of chlorine gas, 35-2, its formula is found to be C12.

The atomic weight of chlorine. — By a careful determination of
the limiting density (p. 147) of hydrogen chloride, F. W. Gray and
P. F. Burt (1909) found the molecular weight to be 36-187 (H = 1).
Hence, the atomic weight of chlorine = 36-187 — 1 = 35'187. By
decomposing the gas with heated aluminium they found that 2 vols.
gave 1-0079 vols. of hydrogen at S.T.P.

The gravimetric composition of hydrogen chloride was directly
determined by Dixon and Edgar (1905), who burnt pure hydrogen
from a weighed palladium bulb, in pure chlorine from a bulb of
liquid chlorine prepared by the electrolysis of silver chloride, and
passed into a previously evacuated glass bulb (Fig. 133), the gases
being ignited by a spark. The hydrogen chloride was absorbed in
water in the bulb, and the residual hydrogen (used in excess) pumped
out. The value Cl = 35'189 was found. Edgar (1908) omitted
the water (which gave a little oxygen when chlorine was used in
excess), and condensed and weighed the dry hydrogen chloride in a
nickel-plated steel bomb, which was placed in liquid air. The hydro-
gen, chlorine, and hydrogen chloride were all weighed and the
synthesis was therefore complete. He found Cl = 35'187, which
is the accepted value.

234

INORGANIC CHEMISTRY

CHAP.

Union of hydrogen and chlorine under the influence of light. —

A mixture of practically equal volumes of hydrogen and chlorine,
containing a minute trace of oxygen, is obtained by the electrolysis
of concentrated hydrochloric acid (p. 231). After the electrolysis
has proceeded for some time, the gas is passed through a series of
very thin glass bulbs (Fig. 134), the whole operation being performed

in a dark room
lighted by a ruby
lamp. The bulbs
are then separated
and closed by
pieces of glass rod
inserted into the
pieces of rubber
tubing. They are
preserved in a dark
box.

Water

C/2

FIG. 133.— Atomic Weight of Chlorine by direct union of -™ HAT*

Chlorine and Hydrogenf EXPT. 110.— If a

bulb, protected by

a screen of plate glass (Fig. 135), is exposed to the light of burning mag-
nesium ribbon, a sharp explosion occurs, and the glass is shattered.

The action of light in bringing about the union of hydrogen and
chlorine is a case of photochemical catalysis. Heat is evolved in
the reaction,
hence the ac-
tion of light
consists only
in initiating the
reaction, which
when once
started goes on
spontaneously
(cf. p. 695).
The action of
light in this
case is called a
trigger-effect.

FIG. 134. — Filling Glass Bulbs with a Mixture of Chlorine and Hydrogen.

EXPT. 111. — Break off the tip of a bulb of mixed gases under potass-
ium iodide solution. The latter is coloured brown, owing to liberation
of iodine : 2KI + C12 = 2K -\- I2, and the liquid rises and half fills the
bulb. If the latter is now depressed in water, and the upper capillary
broken off, the escaping gas may be ignited : it is hydrogen. Expose
another bulb to diffused daylight for a few hours. The colour of the
chlorine disappears. Break off the tip of a capillary under mercury.

xiii COMMON SALT. HYDROCHLORIC ACID. CHLORINE

235

No gas bubbles out, arid no mercury enters. Pour some water coloured
bine with litmus over the mercury, and raise the bulb so that the
capillary enters the water. The latter fills the bulb and its colour
changes to red. Thus 1 vol. of hydrogen -f 1 vol. of ^chlorine = 2 vols.
of hydrochloric acid.

Pringsheim (1887) found that if the mixed gases were carefully
dried with phosphorus pentoxide before passing into the bulb, and
the latter exposed to magnesium light, there was no explosion, but
only a dull click. The bulb became very hot and the gases were
found to have combined completely. (The perfectly dry gases
can be exposed
to sunlight for
several days
without com-
plete combina-
tion occurring,
and without ex-
plosion.) Dixon
and Harker
(1890) found
that the velocity
of the detona-
tion wave
(p. 729) in care-
fully dried hy-
drogen and
chlorine was

in tne moist gas

it was only

1770 m. per sec. Moisture, although assisting the initiation of the

reaction, therefore appears to retard it once it has begun.

J. W. Draper (1843) investigated and confirmed an effect noticed
by Dalton (1809), that a mixture of hydrogen and chlorine did not
begin to contract at once when exposed over water to diffused day-
light. There was an initial " hesitation," called the period of
photochemical induction, or Draper effect. Bunsen and Roscoe
(1857-62) used the apparatus shown in Fig. 136, called an
actinometer, to investigate the reaction. The mixed gases were
confined in the half-blackened flat bulb i by chlorine water. On
exposure to light, contraction occurred, the HC1 formed dissolving,
and the rate of combination could thus be estimated by the move-
ment of the thread of liquid in the horizontal tube k. It was found that
the rate of combination was proportional to the intensity of the light.
These experimenters also noticed the photochemical induction period.

FIG. 135. — Explosion of a Mixture of Hydrogen and Chlorine by
exposure to strong light of burning Magnesium.

236 INORGANIC CHEMISTRY CHAP.

Burgess and Chapman (1904) showed that the period of photo-
chemical induction was not really peculiar to the reaction H2 -f- C12 =
2HC1, but was due to traces of impurities, ammonia or nitrogenous
organic matter, .in the water used to confine the gases. If this
water was first boiled with chlorine, these substances were destroyed,
and the gases then began to combine the instant they were exposed
to light. Traces of oxygen also give rise to a period of induction.
The cause of the induction period is not yet clear.

If moist chlorine is exposed to light, there is a momentary expansion,
due to the heat given out in the reaction : 2C12 + 2H2O = 4HC1 + O2
(Budde effect, 1871). It is not exhibited by chlorine dried with P2O5.

The properties of hydrogen chloride. — Hydrogen chloride prepared
from sodium chloride and sulphuric acid is never perfectly pure, but
contains traces of sulphuretted hydrogen. The pure gas is best
prepared by the action of water on silicon tetrachloride (p. 749) :
SiCl4 -f 2H2O = Si02 + 4HC1. Hydrogen chloride is formed by
the action of concentrated sulphuric acid on many metallic chlorides,
such as those of sodium, potassium, ammonium, calcium, and

magnesium — in general,
any chlorides which form
readily soluble sulphates.
Lead chloride, silver

FIG. 136 .— Actinometer of Bunsen and Roscoe. chloride, and mercUTOUS

and mercuric chlorides

are acted upon only with difficulty. Hydrogen chloride is also
formed by the action of water on the chlorides of silicon, aluminium
(2A1C13 + 6H20 - 2A1(OH)3 + 6HC1), phosphorus, and boron.

The normal density of the gas is 1-63915 gm. per litre. When
very strongly heated the gas is slightly dissociated into its elements :
2HC1 ±=; H2 -f- C12. At 1537°, the dissociation amounts to only
0-274 per cent. (cf. steam, p. 212). The gas is also decomposed to
a slight extent by radium emanation.

When hydrochloric acid gas is passed through a U-tube cooled in
liquid air, it condenses to a snow-white, crystalline solid, which
melts at— 111-4° to a colourless liquid, of density 1-184 at the
boiling point, — 83-4°. The perfectly dry liquid is without action
on zinc, iron, magnesium, quicklime, and some carbonates, all of
which are readily dissolved by the aqueous acid, but it readily
dissolves aluminium with evolution of hydrogen : 2A1 -f- 6HC1 =
2A1C13 -f 3H2. The liquid expands on heating, between — 80° and
-f- 30°, more rapidly than a gas. The critical temperature of
hydrogen chloride is 52-3° ; the critical pressure is 86 atm.

Hydrochloric acid is an essential constituent of the gastric juice,
occurring to the extent of 0-2 to 0-4 per cent, under normal con-

xin COMMON SALT. HYDROCHLORIC ACID. CHLORINE 237

ditions. It is derived in some way, not understood, from the salt
taken with the food.

If a test-tube containing concentrated hydrochloric acid is cooled
in liquid air, the acid becomes very viscous and then solidifies, with
considerable contraction, to a glassy mass.

On exposure to moist air, the concentrated acid, or the gas, fumes.
This is due to the attraction of atmospheric moisture to produce
a solution which has a lower vapour pressure than water, and is
therefore deposited in the liquid state in the form of small drops.
In air dried with sulphuric acid, hydrochloric acid does not fume.

Distillation of hydrochloric acid. — When aqueous hydrochloric acid
containing 20-24 per cent, of HC1 is distilled, under 760 mm. pressure,
the acid passes over completely without change of composition, as
though it were a pure compound. If a weaker acid (e.g., 15 per
cent.) is taken, a more dilute acid passes over into the receiver
until the residue in the retort contains 20-24 per cent, of HC1,
whereas if a stronger acid (e.g., 30 per cent.) is distilled, it loses
hydrogen chloride gas with a little moisture until the same 20-24
per cent, acid is left. In both cases the residual acid of 20-24 per
cent. HC1 then proceeds to distil off without change of composition.
Since the composition remains constant during distillation, the
vapour has the same composition as the liquid, hence the boiling
point (110°) remains constant. This is the maximum boiling
point for the aqueous acid ; both weaker and stronger solutions
boil at lower temperatures. The relative numbers of mole-
cules of HC1 and H20 in the liquid of maximum boiling point

are ^M : Z?^? = 1 : 9-94, or 1 : 10 very nearly. Hence -Bineau
36-2 18

concluded that the liquid, which certainly seems to behave on
distillation like a pure substance (p. 3), was a chemical compound,
HC1,10H20. If this is the case, and it volatilises undecomposed,
its vapour-density should be | (36-5 + 180) = 108-25 ; actually it
was found to be only about 10, showing that the vapour was a mixture :

!L~5 =10. It is still possible, however, that the liquid is a com-
pound. This was negatived by the experiments of Roscoe and Ditt-
mar (1860), who carried out the distillation under various pressures,
and found that the concentration of the acid of maximum boiling point
decreases with the pressure :

Pressure mm. Hg . . 50 700 760 800 1800

Per cent. HC1 in max. b. pt. acid 23-2 20-4 20-24 20-2 18-2

The composition of a compound would be independent of the pressure
over a certain range (possibly limited). It is therefore improbable that

238

INORGANIC CHEMISTRY

CHAP.

HC1,10H,,O exists even in the liquid ; the maximum boiling point acicl
is a solution, and the composition at 760 mm. agrees approximately
with a chemical formula only by accident. It may even then be said
that the liquid is perhaps a compound which is broken up on heating,
but exists at lower temperatures. By passing HC1 gas into the con-
centrated aqueous acid at — 23°, Pierre and Pouchot did obtain a
crystalline hydrate, decomposing on warming, but it was HC1,2H2O
(m. pt. — 18°), not HC1,10H2O. Rupert (1907) obtained the hydrate
HC1,H2O. There is, therefore, no evidence for the existence of a
hydrate HC1,10H2O boiling at 110°.

The manufacture of hydrochloric acid and chlorine. — On the

large scale, hydrochloric acid is made by the action of fairly
concentrated sulphuric acid on common salt (saltcake process). The

FIG. 137.— Saltcake Muffle Furnace.

acid may be mixed in the gaseous state with air, and the mixture
passed over a heated mass containing copper salts, which acts as a
catalyst : 4HC1 + O2^± 2H20 -f 2C12 (Deacon process). The gas
may also be condensed in water in towers, and the solution
(spirit of salt) decomposed by heating with manganese dioxide :
4HC1 + Mn02 = 2H2O + MnCl2 + C12 (Weldon process). Large
quantities of chlorine are now prepared directly from common salt
by electrolysis (p. 296).

The saltcake process. — The first step in the manufacture of caustic
soda by the Leblanc process (p. 777) is to decompose common salt
with sulphuric acid, with the production of sodium sulphate,
Na2S04, known as saltcake. The reaction is carried out in two
The first stage, which proceeds at lower temperatures,

xin COMMON SALT. HYDROCHLORIC ACID. CHLORINE

239

leads to the formation of acid sodium sulphate : NaCl -f H2SO4 =
Xn HS04 -f- HC1. The second stage is carried out by heating this acid
sulphate with common salt, at a dull red heat, when all the acidic
hydrogen is expelled : NaHSO4 -f NaCl = Na2SO4 + HC1.

The operation is carried out in the saltcake pan and muffle furnace
shown in Fig. 137. Half a ton of coarse-grain salt is charged into
the large hemispherical cast-iron saltcake pan, A, and an equal
weight of sulphuric acid, sp. gr. 1 -7, run on. A copious evolution of
hydrochloric acid occurs,
the gas being led off
through p. When this
slackens, the pan is heated
by flue gases admitted by
means of the dampers, /t
and /2. When the first
reaction is completed, the
pasty mass is raked into
the closed box, or muffle,
B, of firebrick, heated
externally by flames from
the fireplace, C, which
functions as a gas pro-
ducer (p. 705). The rest
of the hydrochloric acid
passes out through the
pipe d. Saltcake is left
in the muffle. A modern
furnace produces 85 tons
of saltcake per week.

Hydrochloric acid
towers. — The absorption
of hydrochloric acid gas
in water, which is carried
out in the laboratory in
Woulfe's bottles, is
effected on the large scale
in absorption towers, intro-
duced by Gossage in 1836.
The gas coming from the saltcake furnaces is cooled by passing
through a battery of cast-iron pipes (which are not attacked by
the gas if the temperature is kept above the point of condensation
of the accompanying moisture), and then passes to the base of a
tower 60 ft. high, composed of sandstone slabs boiled in tar and
clamped together with iron bands, which is packed with lumps of
hard coke (Fig. 138). A shower of water is run down, and the
hydrochloric acid is almost completely absorbed. To produce strong

Fm. 138.— Absorption Towers for Hydrochloric Acid.

240

INORGANIC CHEMISTRY

acid (about 33 per cent. HC1) the liquid is recirculated over the
coke packing by acid-pumps of stoneware or ebonite. Efficient
absorption depends chiefly on keeping the tower cool, and presenting
a large wetted surface to the gas. The latter is provided by the
irregularly-shaped lumps of coke, which retains water in its pores.

The manufacture of chlorine by the Weldon process.— Chlorine was
formerly made, for producing bleaching liquor, by Berthollet's
process, in which salt was decomposed by manganese dioxide and
sulphuric acid in stoneware jars heated in a water-bath. In 1836,
however, Gossage began to condense the hydrochloric acid, evolved in
the decomposition of salt with sulphuric acid, in towers. Since large
quantities of salt were decomposed in the manufacture of alkali by
the Leblanc process (p. 777), hydrochloric acid became cheap, and was
used as a source of chlorine by treating it with manganese dioxide.

The operation is carried out, on a small scale, in chlorine stills,

-"i-<- cv-r r.^;.y>v TV '*•?"£". "

FIG. 139.— Chlorine Still.

made of flagstones bound together, and having a false bottom A on
which the lumps of manganese dioxide (pyrolusite) rest (Fig. 139).
Hydrochloric acid from the Gossage towers is run on the manganese
through the pipe B, with a liquid seal below, and the still is heated
by admitting steam cautiously from a stoneware column, C.
Chlorine is evolved through the pipe, Z), and deposits moisture in
the pot shown. The residual liquid in the still contains manganous
chloride, ferric chloride (from impurity in the pyrolusite), and a
fairly large amount of undecomposed hydrochloric acid.

In 1837 Gossage attempted to recover the manganese from this
liquor, by precipitating it with the theoretical amount of lime :
MnCl2 + Ca(OH)2 = CaCl2 + Mn(OH)2. By blowing air through
the manganous hydroxide, Gossage hoped to convert it into mangan-
ese dioxide, which could be used again : 2Mn(OH)2 -f 02 =
2Mn02 -f- 2H20. He found, however, that the oxidation was very
incomplete, and Volhard later pointed out that this was due to
the acidic character of manganese dioxide. The latter combined

xin COMMON SALT. HYDROCHLORIC ACID. CHLORINE 241

with the nianganous oxide, which is basic, to form Mn(XMnO2, or
Mn20}, which is very stable. In 1866 Walter Weldon, working at
Gamble's alkali works at St. Helens, discovered how to make the
Gossage process succeed, and he devised a method for the recovery
of the manganese which was for a long time in extensive operation.
It is known as the Weldon process. Weldon found that if the precipi-
tation of the manganese liquor is carried out in presence of 30-40
per cent, excess of lime, then on blowing air through the mixture the
nianganous oxide is completely oxidised to the dioxide, the latter
combining with the lime to form the compounds CaO,Mn02 and
CaO,2Mn02. Lime is a stronger base than MnO, and prevents the latter
forming a compound with the Mn02 and thus escaping oxidation.

An elevation of a Weldon plant is shown in Fig. 140. The acid
manganese liquor from the stills is neutralised in the well, A, by
agitation with limestone, and the ferric hydroxide precipitated is
allowed to settle out in the tanks, B. The liquor is then pumped into
the oxidiser, (7, consisting of a large cylindrical iron tank, where it is
treated with the requisite excess of milk of lime. The liquor is
heated to 60° by blowing steam into it, and a powerful blast of air
is forced through it from a blowing engine. The compound
CaO,Mn02, or calcium manganite, is precipitated. More still-liquor
is run in, and the blowing continued, when some of the compound
CaO,2MnO2 is formed. The suspension from the oxidiser is then
run into the settling-tanks, Z), where a thin black mud, called
Weldon mud, settles* out. The clear liquor, containing calcium
chloride, is drawn off and thrown away ; the mud (calcium mangan-
ite) run down into the octagonal stone chlorine stills, E, where it
is treated with hydrochloric acid and steam, producing chlorine, and
manganese liquor. The latter goes through the Weldon process
repeatedly, as described, but fresh manganese dioxide must be added
to replace losses. The reactions in the Weldon process are as follows :

1 . Stills (E) : (a) MnO2 + 4HC1 =MnCl2 + C12 -f H2O (fresh pyrolusite).

(6) CaO,2MnO2 -f 10HC1 = 2MnCl2 + CaCl2 + 2C12 + 5H2O
(mud).

2. Neutralising tank ( A) : 2FeCl3 -f 3CaCO3 + 3H,O = 2Fe(OH)3

-|- 3 Ca012 + 3CO,.

3. Oxidiser (C):

(a) adding lime : Mn012 + Ca(OH)2 = Mn(OH)2 + CaCl2.

(b) air-blowing : 2Mn(OH), -f- 2Ca(OH)2 -f O, = 2CaO,MnO2 -f-

4H2O.

2 CaCl3 + 4H2O.

A considerable amount of chlorine is wasted in this process as calcium
chloride.

242

INORGANIC CHEMISTRY

CHAP.

The Deacon process. — The oxidation of hydrochloric acid gas by
atmospheric oxygen in the presence of a catalyst (ExpT. 85) was
applied by H. Deacon and F. Hurter in 1868 as a technical process
for the manufacture of chlorine : 4HC1 + O2 = 2H2O + 2C12.
The reaction is reversible ; the reverse reaction was described on

FIG. 140— Weldon Chlorine Plant.

p. 160. The hydrochloric acid gas, mixed with air, was passed over
broken bricks soaked in copper sulphate solution and heated to
about 500°. Although chlorine was evolved at first, the reaction
soon stopped, and the process, on which great hopes were based,
promised to become a complete failure. Hasenclever, in 1883,
improved the method, and chiefly in his hands the Deacon process

xiii COMMON SALT. HYDROCHLORIC ACID. CHLORINE 243

was converted into a successful technical operation, which almost
completely displaced the older and wasteful Weldon process.

Hasenclever found that the contact mass, impregnated with
copper salt, lost its activity slowly in any case, and had to be
replaced from time to time. He therefore used a container, called
a decomposer, consisting of an upright iron cylinder, 12-15 ft. wide,
containing a ring of broken bricks, previously dipped into a solution
of cupric chloride so as to contain 0-6-0-7 per cent, of copper in
the mass, supported by iron shutters, and divided into six compart-
ments, one of which can be emptied and refilled with fresh contact
mass every fortnight (Fig. 141). The
mixture of air and hydrochloric acid
gas, 1 vol. of HC1 to 4 vols. of air, is
passed by a hot Roots' blower through
a set of iron pipes heated in a furnace,
called a preheater, where its temperature
is raised to 450°. The gases then pass
to the converter, which is kept at this
temperature by the hot flue gases from
the preheater. About two-thirds of
the HC1 is decomposed, and the rest is
washed out of the gas in a coke-tower
with water. The dilute chlorine, con-
taining 5-10 per cent, of chlorine,
diluted with nitrogen, is then dried in
a sulphuric acid tower, and used in
making bleaching powder (p. 376). A
diagram of the apparatus is shown in
Fig. 142. Hasenclever 's main improve-
ment, however, was the preliminary
purification of the hydrochloric acid
gas. He collected the crude gas from
the saltcake furnaces in a Gossage
tower, and then ran the aqueous acid
in a slow stream into concentrated

sulphuric acid, blowing out - the hydrochloric acid gas with a
current of air. Poisoning of the catalyst was then very much
reduced.

The reaction of the Deacon process is reversible, and the HC1
cannot be completely decomposed. The proportion of decomposition
diminishes with rise of temperature, so that the process must be worked
at the lowest possible temperature. Below about 350°, however,
there is practically no decomposition, and the reaction only. becomes
sufficiently rapid at 425-450°. We therefore have two opposite
conditions to satisfy, (i) the yield of chlorine, which decreases with
rise of temperature ; (ii) the speed of the reaction, which increases

R 2

Fia. 141. — Deacon Converter.

244

INORGANIC CHEMISTRY

CH. XIII

with rise of temperature. A technical balance is struck at about
450°, when about two-thirds of the HC1 is decomposed.

When the Deacon process got into complete techni
cal operation, and displaced the Weldon method, it
found iiself threatened by a new competitor, which

FIG. 142.— Diagram of Deacon Chlorine Plant.

•

will doubtless in time oust the contact process from the field,
is the electrolytic process, described in Chapter XVI.

This

EXERCISES ON CHAPTER XIII

1. Describe the occurrence, manufacture, and properties of common
salt. How may (a) hydrochloric acid, (b) chlorine be prepared from it ?

2. What weight of chlorine would be obtained by decomposing 100
gm. of common salt with manganese dioxide and sulphuric acid ?
What volume of water at 15° would be required to dissolve this chlorine ?

3. What experiments would you make in order to demonstrate (a) the
solubility of hydrogen chloride in water, (b) that hydrogen chloride
contains half its volume of hydrogen, (c) that hydrogen and chlorine
combine explosively when exposed to strong light, (d) that hydrochloric
acid contains chlorine, (e) that the bleaching action of chlorine depends
on the presence of water ?

4. How are pure hydrogen chloride and chlorine prepared ? Describe
their properties.

5. Describe the manufacture of hydrochloric acid from common salt.
How is chlorine prepared from hydrochloric acid on the large scale ?

6. What is the action of concentrated hydrochloric acid on
(a) manganese dioxide, (b) lead dioxide, (c) potassium permanganate,
(d) potassium dichromate, (e) barium peroxide ? Give equations.

6. What views have been held as to the nature of chlorine ? Why
is it now supposed to be an element ? Describe, on the assumption that
chlorine is oxymuriatic acid, (a) the action of manganese dioxide on
hydrochloric acid, (6) the action of hydrochloric acid gas on heated lead
oxide, (c) the union of sodium and chlorine.

7. Describe the experiments which have been made on the union of
hydrogen and chlorine under the influence of light. How would you
prove, by making use of this reaction, that the chemical action of light
is proportional to its intensity ?

CHAPTER XIV

VALENCY AND THE STRUCTURE OF COMPOUNDS

Valency. — Hydrogen compounds exist in which one atom of an
element is combined with one, two, three, or four atoms of hydrogen :

HC1 H20 H3N H4C

Hydrochloric acid. Water. Ammonia. Methane.

The atoms of chlorine, oxygen, nitrogen, and carbon are capable of
uniting with one, two, three, and four atoms of hydrogen, respectively.
No compound of hydrogen, except hydrazoic acid, HN3, is known
containing more than one atom of an element combined with one
atom of hydrogen, and the latter is therefore taken as the standard
of combining capacity or valency. The valency of an element is
measured by the number of hydrogen atoms which unite with one
atom of that element. Thus chlorine, oxygen, nitrogen, and carbon
are univalent, bivalent, tervalent, and quadrivalent respectively.

Since chlorine is univalent, it may be used instead of hydrogen
in determining the valencies of elements. The valencies of elements
thus found are the same as those referred to hydrogen, but quin-
quevalent and sexivalent elements are now included :

C120 C13N C14C C15P C16W

Chlorine Nitrogen Carbon Phosphorus Tungsten

monoxide. trichloride. tetrachloride. pentachloride. hexachloride.

In the compounds chlorine monoxide, C12O, and calcium chloride,
CaCl2, oxygen and calcium are bivalent. When, therefore, calcium
and oxygen combine, we should expect them to do so atom for atom,
since each of the combining atoms has a valency of two units. This
is the case ; calcium oxide, or quicklime, has the formula CaO. The
valency of calcium may also be inferred from the fact that it can
displace two atoms of hydrogen, and occupy their place : 2HC1 -f-
Ca = CaCL + H2. If chlorine is passed over strongly heated lime,

245

246 INORGANIC CHEMISTRY CHAP.

one atom of oxygen is displaced by two of chlorine : 2CaO -f- 2C12 =
2CaCl2 + 02.

There is obviously a close relation between the atomic weight
and equivalent of an element and its valency. The equivalent is the
weight of an element which combines with or displaces unit weight
of hydrogen. But the valency is the number of unit weights
(atoms) of hydrogen which combine with, or are displaced by, one
atomic weight of the element, hence :

Atomic weight = Equivalent X Valency,

or Valency = Atom^weight
Equivalent

Valency volume. — The simplest conception we can form of the
displacement of a group of w-univalent atoms from the molecule of
a compound by one n-valent atom, e.g., Al -j- H3P04 = A1P04 + 3H,
is that the 7i-valent atom occupies the space previously taken up by
the ?i-univalent atoms. This representation will, it is true, be
limited to solid compounds, because it is only in these that the
atoms are in close proximity. Barlow and Pope (1906) found that
this relation is true in a number of cases ; they regard the volume of
an atom as proportional to its valency, provided the atoms are
arranged in the condition of closest packing, although arbitrary
assumptions have often to be made which will require further justi-
fication.

Oxygen compounds. — If we examine a series of oxygen compounds :

Na20

(Ca202)

A1203

(C204)

NA>

(S206)

C120,

(Os208)

Sodium

Calcium

Alumi-

Carbon

Nitro-

Sulphur

Chlorine

Osmium

mon-

oxide

nium

dioxide

gen

trioxide

hept-

tetr-

oxide

pent-

oxide

we see that two additional higher valencies, 7 and 8, appear.
Chlorine is septavalent, and osmium is octovalent, in their highest
oxides. (The formulae of CaO, C02, S03, and Os04 have been
doubled for clearness.) The number of atoms of oxygen combining
with two atoms of an element is a measure of the valency of the
latter, since oxygen is bivalent. The valency of 8, shown in the
oxygen series, and there only in the compounds osmium and
ruthenium tetroxides, Os04 and Ru04, is the highest value ever
exhibited. The inactive gases argon, helium, etc., form no com-
pounds with any elements, and their valency is zero. We have
therefore, in all, nine valencies, shown by various elements, viz.,
0, 1, 2, 3, 4, 5, 6, 7, and 8.

Classification of elements according to valency. — We can classify
all the elements in eight groups (if we exclude zero valency), accord-

xiv VALENCY AND THE STRUCTURE OF COMPOUNDS 247

ing to their valencies. The same element may fall into several
groups, since it has been shown that the valency may be different,
according as the element is combined with hydrogen (HC1,H2S), or
with oxygen (C12O7,S03). These groups are as follows :

0. Zero-valent elements : inactive gases, radioactive emanations
(p. 463).

1. Univalent elements : hydrogen, halogens, alkali-metals, silver,
nitrogen in nitrous oxide, N2O, mercury in mercurous compounds
(HgCl, Hg2O), copper in cuprous compounds (CuCl, Cu2O), gold in
AuCl.

II. Bivalent elements : oxygen, nitrogen in nitric oxide, NO,
alkaline-earth metals (Ca, Sr, Ba), magnesium, zinc, cadmium, mercury
in mercuric compounds (HgCl2, HgO), copper in cupric compounds
(CuCl2 ,CuO), tin in stannous compounds (SnCl2, SnO), lead in plumbous
compounds (PbCl2, PbO), iron in ferrous compounds (FeCl2, FeO),
sulphur in H2S, SaCl2.

A large number of elements are seen to be bivalent.

III. Tervalent elements : aluminium, boron, nitrogen in NH3 and
NC13, iron in ferric compounds (FeCl3, Fe2O3), phosphorus in PH3 and
PCL, arsenic in AsH3, AsCl3, As2O3, antimony, bismuth, gold in AuCl3.

IV. Quadrivalent elements : carbon, silicon, nitrogen in nitrogen
dioxide, NO2, lead in plumbic compounds (PbCl4, PbO2), sulphur in
SO2, tin in stannic compounds (SnCl4, SnO2), platinum.

V. Quinquevalent elements : nitrogen, phosphorus, arsenic, and
antimony in higher halogen or oxygen compounds (N2O5, PC16, As2O6,
SbCl5), manganese in manganese trioxide, MnO3, and manganates,
K2MnO4.

VI. Sexivalent elements : sulphur in SF6 and SO3, tungsten in WC16.

VII. Septavalent elements : chlorine in C12O7, iodine in KIO4,
manganese in Mn2O7, KMnO4.

VIII. Octovalent elements : osmium in OsO4 and OsF8, ruthenium
in RuO4.

Variable valency. — An element may exhibit a variable valency

either in its compounds with the same element :

PC13 (3) S02 (4) N203 (3)

PC15(5) S03(6) N205(5)

or in its compounds with different elements :

NH3 (3) PH3 (3) SH2 (2)

N205 (5) P205 (5) SF6 (6)

It will be noticed that the valency of an element is usually either
odd or even, but exceptions are known, e.g., WC15 (5), WC16 (6),
and NH3 (3), NO (2).

248 INORGANIC CHEMISTRY CHAP.

The lowest valency is always shown in the hydrogen compounds,
and the highest valency in the oxygen compounds.

If an element, especially a metal, forms two or more series of
compounds in which it has different valencies, the properties of the
compounds in these series are usually totally different. As an
example, we may compare the properties of the ferrous (bivalent
iron) and ferric (tervalent iron) compounds :

Ferrous sulphate. Ferric sulphate.

1. Green crystals, FeSO4,7H2O. 1. White powder, Fe2(SO4)3.

2. Greenish-white precipitate with 2. Brown precipitate with ammo-

ammonia, Fe(OH)2. nia, Fe(OH)3.

3. Bluish -white precipitate with pot- 3. Dark-blue precipitate with pot-

assium ferrocyanide. assium ferrocyanide.

4. Deep blue precipitate with pot- 4. No precipitate, but dark brown

assium ferricyanide. colour, with potassium ferri-

cyanide.

5. No coloration with ammonium 5. Blood-red coloration with am-
f- thiocyanate. monium thiocyanate.

6. Double salt, K2SO4,FeSO4,6H2O, 6. Double salt, K2SO4?Fe2(SO4)3,

pale green crj^stals. 24H2O, amethyst -coloured

crystals.

Unless we knew that ferrous and ferric sulphates were both salts
of the same element, iron, the first convertible into the second
by boiling with nitric acid, these tests would reasonably lead us to
conclude that we had to do with salts of two entirely different
elements.

Compounds of an element in which it has a particular valency may
resemble compounds of another element of the same valency more
closely than they resemble other compounds of the first element with
a different valency.

Thus, silver, mercurous, and cuprous chlorides are all white,
sparingly soluble solids : AgCl, HgCl, CuCl. Mercuric and cupric
chlorides are soluble, and the latter has a green colour. CuCl is
therefore more analogous to HgCl and AgCl than to CuCl2. Bivalent
lead and tin compounds resemble each other more closely than com-
pounds of bivalent lead resemble compounds of quadrivalent lead,
or than compounds of bivalent tin resemble those of quadrivalent
tin. Quadrivalent lead and tin are closely analogous :

SnCl2 white, crystalline solid SnCl4 colourless, fuming liquid

PbCl2 „ „ „ PbCl4 yellow, „ ' „

In order to distinguish between the various valencies of an

xiv VALENCY AND THE STRUCTURE OF COMPOUNDS 249

element, a Roman numeral representing the valency may be written
over the symbol of the atom :

I II III IV V VI VII VIII

H O N C P S Cl Os.

Thus, H2SO4 indicates that sulphur in sulphuric acid is sexi-
valent.

Structural formulae. — We may form a crude picture of the
combination of atoms by assuming that each atom possesses one or
more hands, which we represent by bonds, or straight lines drawn
from the symbol of the atom, as many bonds being drawn as the
atom possesses valencies :

_O N— — C— —

H- — O N C Pf =Sf — Clf= 10s

In chemical combination these bonds unite in pairs, i.e., the
hand on one bond grasps the hand on another bond :

H— xx— H H— x— 0— x— H

In writing the formulae of the compounds, the pairs of mutually
satisfying bonds are contracted to single bonds :

Hx /H H

H— H H— 0— H NSK

H— C— H

.0 ^O

\C1 X0 \0

Multivalent atoms are capable of linking with each other, by
utilising one or more bonds on each atom ; the remaining valencies
are free to attach other atoms :

H H N^

H-C C-H .0

I I N^

H H X0

Ethane Nitrogen tetroxide

Such formulae as we have just been using are called structural

formulae ; they are supposed to represent the way in which the
atoms are united, but not their actual positions, in the molecules.

250 INORGANIC CHEMISTRY CHAP.

Thus, the formula for nitrogen tetroxide shows that two nitrogen
atoms are united by a single bond, and each nitrogen atom is directly
united with two oxygen atoms, in each case by a double bond. But
all the following formulae, which would correspond with different
positions in the molecule :

N/° N/° 0=N=0 N/>

"V -L^<\ I ^"V

I ° °V ' ° 0-N-O 1 °

,0 >N 0=N=0

N^ (K

express in this case, where the nitrogen atoms are united by a single
bond, exactly the same thing, and are therefore really the same.

Saturated and unsaturated compounds. — In some cases it is
assumed that two or more valencies of an atom of an element can
unite with a corresponding number of an atom of the same element :

H H

I |

1. Ethane, H — C — C — H, single bond or linkage, between carbon

atoms.

H H

. I I

2. Ethylene, C=C, double bond, or linkage, between carbon

I atoms.

H H

3. Acetylene, H — C=C — H, treble bond, or linkage, between carbon

atoms.

Such double and treble bonds are often represented by dots, to
save space in printing, thus: H3C-CH3, H2C:CH2, HC:CH,
which are usually written CH3-CH3, CH2:CH2, CHjCH.

The propriety of this mode of representation is shown by the fact
that the molecules of compounds with multiple bonds are unsatu-
rated, i.e., they can add on other atoms to form saturated compounds :

+H2 +H2

GH;CH->CH2:CH2 -> CH3'CH3.

The multiple linkages therefore contain latent bonds, each linkage
when broken giving two available bonds :

HfeCH -> HC=CH ^> H2C=CH2 -> H2C— CH2
\/ \/

H2C— CH2 ^l H3C-CH3 (saturated).

xiv VALENCY AND THE STRUCTURE OF COMPOUNDS 251

It was formerly assumed, and supported by the authority of
Kekule, that variable valency is really always due to latent bonds.
Thus, phosphorus was supposed to be always quinquevalent, but
in compounds in which it is apparently tervalent two bonds are
latent or unsaturated :

Cl Cl

/Cl CL |

< \P

xa A

Support was lent to this idea by the circumstance, pointed out by
Odling, that when the valency of an element changes, it usually
does so two units at a time. This, however, is not always the case
(cf. p. 247). The discovery of the stable compound PF5 (p. 623)
entirely vitiated the hypothesis of constant valency.

Valency of radicals. — The conception of valency may be applied
not only to the atoms of the elements, but also to the radicals, or
groups of atoms which take part as a whole in chemical reactions.
Thus, in the hydrocarbons ethane, ethylene, and acetylene, we
recognise the uni-, bi-, and ter-valent radicals — CH3 (methyl),

(methylene), and -~)CH, respectively. In the same way an
inspection of the table :

g H2S04 H3PO4 Na3P04

NaN03 K2S04 NaH,PO4 NaNH4HPO4

NH4N03 (NH4)2S04 Na2HP04 (NH4)3P04

leads to the recognition of the following radicals in the compounds :
— N08, J>S04, ^P04, and — NH4.

If we know the valencies of the elements and of common radicals,
we can at once write down the formulae of all the salts formed from
them.

It is usually most convenient to remember the formulae of a few
typical compounds, from these to deduce the valencies of the elements
or radicals, and thence to write down the formula of the compound
required.

Thus, if we wish to write down the formula of aluminium sulphate,
we remember the formulae A1C13 and H2SO4. Hence we find that Al
is tervalent arid SO4 bivalent: — Al<, >SO4. In order to satisfy
the valencies of Al by those of SO4, we shall have to take 2A1, i.e., 6
valencies, and 3SO4, "also 6 valencies. No free valencies must be left
over. Hence aluminium sulphate is A12(SO4)3.

252 INORGANIC CHEMISTRY CHAP.

The following table contains the valencies of a few common
elements and radicals. They are arranged into electropositive and
electronegative groups. The former include elements or radicals
which are attracted to the negative pole in electrolysis, the latter com-
prises elements attracted to the positive pole (Chapter XVI). They
are also in the order of the list given on p. 133 ; metals and hydrogen
are electropositive ; oxygen and halogens are electronegative ;
the other elements are sometimes positive and sometimes negative.

In its compounds with hydrogen or metals an element is assumed
to be electronegative ; in compounds with oxygen, halogens, and
sulphur it is electropositive. Negative atoms or groups, especially
those containing oxygen, confer acid properties : HC1,H2S04.

TABLE OF VALENCIES OF ELEMENTS AND RADICALS.

1. Univalent : —

Positive : H, Na, K, Li, Cu(ous), Hg(ous), Ag, (NH4).

Negative : Cl, Br, I, F, (NO3), (OH), (MnO4) (in permanganates).

2. Bivalent : —

Positive : Mg, Sr, Ca, Ba, Fe(ous), Sn(ous), Pb, Cu(ic), Zn, Cd,

Hg(ic), Cr(ous).
Negative : (SO8), (SO4), (CO8), (CrO4), (MnO4) (manganates).

3. Tervalent :—

Positive : Al, Fe(ic), Cr(ic), As(ous), Sb(ous), Bi, N, B.
Negative : N, PO4, AsO3, AsO4.

4. Quadrivalent :—

Positive : Si, Sn(ic), Pb(ic), C (in CCL).
Negative : C (in CH4), Si (in SiH4).

5. Quinquevalent : —

Positive : N (in N2O5), P (in PC15, P2O6).

6. Sexivalent : —

Positive : S (in SF6, SO3), Or (in CrO3).
7 Septavalent :—

Positive : Cl (in C12O7), Mn (in Mn2O7, KMnO4).
8. Octovalent : —

Positive : Os (in OsF8), Ru in RuO4.

Elements of low valency are either distinctly electropositive or
distinctly electronegative (e.g., alkali metals, halogens). This
sharp definition of properties falls off as the valency increases ;
quadrivalent elements have practically no electrochemical character,
and are sometimes weakly positive (CC14), and sometimes weakly
negative (CH4). Elements of valency higher than 4 are all positive.

Molecular compounds. — Saturated molecules often have the
capacity of uniting with each other, although they cannot take up
additional atoms of elements. Thus, hydrofluoric acid, HF, and
potassium fluoride, KF, although both are saturated compounds,
combine to form the salt potassium hydrogen fluoride, KHF2. This

xiv VALENCY AND THE STRUCTURE OF COMPOUNDS 253

salt is readily broken up on heating, into KF and HF, and ^ence it is
usually formulated as KF,HF, as though the separate molecules are
contained in it, and called a molecular compound.

The explanation of the formation of compounds from apparently
saturated molecules is based on the hypothesis of residual valencies.
The free positive valency of potassium is not quite neutralised by the
free negative valency of fluorine when the elements combine atom
for atom ; in order to bring about complete neutralisation, a
fraction of an atom more of fluorine would be required. The addition
of this fraction of an /itom is impossible, hence the KF molecule
exhibits a residual positive valency. The electronegative valency
of fluorine is not entirely neutralised by the positive valency of
hydrogen, hence the HF molecule exhibits a residual negative
valency. These residual valencies are represented by dotted lines
instead of by bonds ; they are less than a unit of free valency as

exhibited by a hydrogen atom : KF . . . and HF . . . The two
residual valencies, although not capable of uniting with a univalent
atom, can unite with each other, forming the molecular compound
KF . . . HF. The constituents of molecular compounds are
usually separated by commas, e.g., KF,HF.

It is usual to recognise three kinds of valencies : (1) free positive
valencies and (2) free negative valencies, exhibited by atoms or
radicals ; (3) residual valencies, exhibited by molecules.

Determination of valency. — The valency of an element in a
particular compound can be determined with certainty only
(i) from compounds containing a single atom of the element, (ii) if
the molecular weight of the compound is known, and thence its
molecular formula.

Silica, SiO2, is a non-volatile solid, and its formula may be SiO2,
Si2O4, Si306, ... or generally, (Si02)n. In SiO2, silicon is quadri-
valent : 0=Si— 0, but in Si204 it is quinquevalent :

CK 0

Silicon is assumed to be quadrivalent because the compound SiCl4 is
volatile, and its molecular weight can be found. The molecular
weight of a substance can also be found in solution, p. 299 ; hence the
compound must be volatile, or must dissolve without decomposition
in a solvent, in order that its molecular weight can be found. Very
complicated formulae of silicates, for instance, are found in chemical
literature, but as the compounds are neither volatile nor soluble, the
structural formulae are guesswork, and have very little scientific
value. The presence of certain groupings of atoms in a compound
mav often be inferred from chemical reactions, and in the case of

254 INORGANIC CHEMISTRY CHAP.

carbon compounds, where the valencies are not usually variable, this
may lead to the structural formula of the compound.

Thus, alcohol lias the empirical formula C2HflO. This is also the
formula corresponding with the molecular weight, deduced from the
vapour density. One of the hydrogen atoms, however, is in a different,
relation towards oxygen from the rest, since the action of phosphorus
pentaehloride, or hydrochloric acid, leads to the formation of the com-
pound ('..Mr/-!, and only one atom of hydrogen is displaced by sodium,
forming < 'J I r,()Na. The group (OH) in the former reaction has therefore
been replaced by the univalent atom Cl, arid since with many other com
pounds the same reaction is exhibited, we assume that one atom of
hydrogen in alcohol is present as a hydroxyl group, so that the formula
must be written C2H6-OH. The compound, C2H6C1, is also produced
by replacing one atom of hydrogen in the hydrocarbon ethane, C2II(;.
by chlorine : C2H6 + C12 = C2H6C1 + HC1.

The following formula; have been found from direct measurements of
vapour densities, in some cases (e.g., AgCl at 1735°) at very high
temperatures :

WC16

NaCl

BeCI,

A1C13 (above 800°)

TiCl4

Nb016

KC1

CrCl2

CrCl3

VC14

TaCl6

Fe012

FeCl3 (at 750°)

GeCl4

MoCl5

RbCl

Cu2Cla

GaCl3

SnCl4

WC15

Hg,Cl2

CsCl

ZnCl2

InCJ3

ZrCl4

Csl

GaCl2

SbCl3

UC14

AgCl

SnCl8

BiCl3

InCl

InCl2

T1C1

HgCI.,

PbCla

The valencies of the corresponding metals, except in the cases of
Cu2012 and Hg2Cl2, are therefore placed beyond doubt. In many
cases the valency has been confirmed by the vapour densities of
volatile organo-metallic compounds :

Zinc methyl Zn(CH3)2. Lead tetraethyl Pb(C2H5)4

in
Aluminium acetonylacetone Al(C5rI7O2)3.

The valency of an element may be determined from the ratio of
the atomic weight to the equivalent, if the atomic weight can be
found. In most cases this requires a knowledge of the molecular
weight of a series of compounds (p. 143), but sometimes the "deter-
mination of the specific heat of the element in the free state may be
used to ascertain the atomic weight. According to Dulong and

XIV VAU'IM'Y AND THK STIMVTl KM (>K ( '( ).\l l'( )l M )S 255

Petit's law (p. 14(>), the product of the specific heat and atomic
weight of a solid clement is constant, and e<|iial to (i-.'l. If tlie atomic
weight is found in this way, and divided by the e(|iii\ alent , deter-
mined by a particular method, the valency of the element is
found.

Thus, the equivalent of y.inc, as determined by the amount of h\ drogon
evolved by the action •>!' I gm. of y.inc on dilute hydrochloric or sul-
phuric acid, is :52'f>. The specific heat of the solid metal is 0-0'
hen L-O the atomic wi^lit of y.inc is approximately 6*3 <4- 0*0955 = (>,r>.
Cut. :>2-f> X 2 — <>">, henco the valency of zinc in the chloride and
sulphate is '_>, and the formula' of t hoso compounds are ZnCLnnd /nSO4.

Causes Of Variation Of valency.- -The valency of an element may
alter as a result of physical or chemical causes. Thus, phosphorus
p< ntachloride, IVIr>, containing (piiiKjuevalent phosphorus, is
decomposed by heat into phosphorus trichloride, I'CI.,, containing
tervalent phosphorus : PC15 ^ PC13 -|- C12. l'('lr) is not a molecular
compound, as it volatilises almost unchanged in an atmosphere of
!'('!... Meivurous oxide is decomposed bv light into mercuric oxide

i 11

and metallic mercury: HgaO = Hg -|- HgO. Phosphine, PH3,
and hydrochloric acid, IK 'I, do not combine at ordinary pressuie,
but under increased pressure they give solid phosphonium chloride
containing quinquevalent phosphorus: PH., | Hri-~ril4(1.

Chemical changes often lead to alteration of valency : e.g., accord-
ing as an excess or deficit of an element or radical is present during
the preparation of the compound (cf. KXPT. 55) : 2Hg (excess) -f- T2 =

2Hgl ; Hg | L (excess) Hgl2- In sonic eases only one compound
is formed under all conditions from the elements, e.g., tin always
forms stannic chloride with chlorine : Sn -f 2C12 = SnCl4, even if
tin is in excess, but the action of hydrochloric acid leads to the
formation of stannous chloride: Sn | 2HT1 -- Sn('l2 -f- H2.
Changes of oxidation and reduction bring about changes of valency :

II IV 111 II

Sn -> Sn() -> SnO., (oxidation); FeaOa •> FeO (reduction). Oxi-
dation leads to increase of valency, reduction to decrease of valency. The

change of a ferrous to a ferric salt, for instance, is also called
oxidation, because the valency of iron in ferric salts is higher than
that in ferrous salts, and the two series of salts may also often be
regarded as derived from a higher and lower oxide of iron, respec-
t i\ ely :

II
ferrous sulphate .. .. FeSO4 FeO,SOs

ill
ferric sulphate .. .. Fe2(SO4)8 1<V,< >:i,:*SO3.

ferrous chloride, I'VCI.,, is said to be " oxidised " to ferric chloride,

256 INORGANIC CHEMISTRY , CHAP.

FeCl3, since an increase in valency results : neither compound
contains oxygen.

By passing chlorine through a green solution of potassium
manganate, K2Mn04, it is oxidised to a purple solution of potassium
permanganate, KMnO4, and the valency of manganese is raised from
6 to 7 :

K— (X vi ^ K— (\ vii ^0

>MnC +Cl-> ;>Mn( + KC1

K— (K X0 (K XO

Removal of an electropositive atom (K) is therefore equivalent to
oxidation ; addition of an electronegative atom is also oxidation :

ii in

FeCl2 -J- Cl —> FeCl3. The reverse changes are equivalent to reduc-
tion.

An example of a complicated oxidation and reduction occurring
simultaneously is the action of ferrous sulphate on potassium perman-
ganate :

VII II II III

2KMnO4 + 10FeSO4 + 8H2SO4 = K,SO4 + 2MnSO4 + 5Fe2(SO4)3-f
8H20.

II ITI

lOFe becomes lOFe, an increase of 10 units of positive valency
(oxidation).

VII II

2Mn becomes 2Mn, a decrease of 10 units of positive valency (reduction).

SUMMARY OF CHAPTER XIV

The valency, or combining capacity, of an element is measured by the.
number of atoms of hydrogen which can combine with one atom of the
element. Valencies measured in this way vary from 1 to 4. Oxygen
is bivalent (H2O), but if we examine oxygen compounds we find that
elements in them can have valencies from 1 to 8. The inactive gases,
since they form no compounds, may be regarded as zero-valent.

Structural formulae are obtained by linking the atoms (or radicals)
so as to satisfy all the valencies in pairs, e.g.,

H\ /H

H -) C— C £-H, or CH3-CHr

H/ \H

Saturated molecules may combine to form molecular compounds ;
these may be assumed to be formed by residual valencies : KF . . HF.

Unsaturated compounds contain latent valencies, represented by
double, or treble, bonds, which can add on univalent atoms in pairs to
form saturated compounds : CH2:CH2 + C12 = CH2C1-CH2C1.

xiv VALENCY AND THE STRUCTURE OF COMPOUNDS 257

EXERCISES ON CHAPTER XIV

1. Explain what is meant by the statement : " the valency of sulphur
in sulphur trioxide is six." What independent evidence is there in
support of this ?

2. How is the valency of an element determined ? Of what use is the
conception of valency in chemistry ?

3. What is meant by constant valency, varying valency, residual
valency, saturated and unsaturated compounds, double linkages,
molecular compounds ?

4. Classify the common elements according to valency. Write down
the formulae of bismuth sulphate, aluminium silicate, barium phosphate,
calcium permanganate, silicon carbide, ferric phosphate.

5. How may the valency of an element be caused to vary ? What
relation does valency bear to oxidation and reduction changes ?

CHAPTER XV

THE MOTION OF MOLECULES

The kinetic theory of gases. — Dalton in 1801 filled two bottles
(Fig. 143), one with hydrogen and the other with carbon dioxide,
and connected them by a long vertical glass tube, the light gas being
above and the heavy gas below. After several hours the gases were
found to have mixed uniformly, as may be shown by opening each
under caustic soda solution and measuring the absorption. Since
the gases have moved in opposition to the force of gravity, this
spontaneous mixing of gases, called diffusion, must be due to the
motion of the molecules of the gases amongst each other. This
motion in the gases is not perceptible to the eye because the
molecules are so very minute.

Similar diffusive motions occur also in liquids, but even more
slowly. If a tall cylinder is filled with water, and a layer of copper
sulphate crystals placed at the bottom (Fig. 144), the salt dissolves,
and a blue solution is formed, with colourless water above. If
the jar is set aside in a room of uniform temperature, to avoid
convection currents, it will be found that the blue colour slowly
rises through the jar, until after several months the colour of the
solution has become uniform.

We are therefore led to assume that the molecules of liquids and
gases are in ceaseless motion, in much the same way as a swarm of
gnats on a summer evening. This mental picture of the condition
of a molecular swarm, as we conceive it to exist in gases and liquids,
is called the kinetic molecular hypothesis (Greek kinetos, motion),
or, more briefly, the kinetic theory.

From the slowness of diffusive motion it might seem that the
molecular speeds must be small. This is not correct, and we shall
see later that the molecules in air, for instance, are flying about
with speeds of the order of a quarter of a mile per second. In the
same way the gnats in the swarm are moving about with consider-
able speeds, although the swarm itself is nearly stationary.

The cause of gaseous pressure. — It was shown by Joule in 1845
that if a gas is allowed to expand into a rigid evacuated vessel, in

258

CH. xv THE MOTION OF MOLECULES 259

such a way that it does no external work, it does not become
appreciably warmed or cooled. Hence no appreciable work has
been done by, or against, possible forces of repulsion or attraction
between the molecules, and we must therefore conclude
that the molecules of gases exert practically no forces on one
another.

The pressure exerted by the gas uniformly over the
walls of the vessel containing it must therefore be wholly
kinetic in origin — in other words it must be caused by
molecular bombardment. On all parts of the surface there
is a ceaseless hail of molecules, which impinge on the
surface and fly off again into the gas. Without going
further into the dynamics of the question one can see that
this molecular bombardment, distributed over the surface,
must appear to our coarse senses as a uniform pressure.

The molecules strike the wall at all angles, from a full
normal blow to a glancing impact, and it is evident that it
is only the component of the velocity perpendicular,
or normal, to the surface which is effective in producing
pressure.

In the gas itself the molecules, since they exert
practically no forces one upon another, will move in
straight lines until they encounter the walls, or one
molecule collides with another. These molecular collisions
will occupy but a small fraction of the
whole time in which the molecule is moving,

because the particles are sparsely dis- v™; 143/
,., T £ • i • 7 i -i Dal ton s

tnbuted, except in highly compressed gases. Experiment

on Gaseous
Thus, 1 c.c. of water gives 1240 c.c. of Diffusion.

vapour at 100° and 760 mm. pressure, so that
less than one-thousandth of the whole space of the
vapour is occupied by the volume of the molecules.
In air at 0-001 mm. pressure, the molecules occupy
only about 1 part in 580 millions of the total space.
The gaseous state of matter is therefore one of con-
siderable attenuation.

The molecules can have all possible speeds from zero
to infinity, but Clerk Maxwell (1859) showed that the
speeds of the vast majority of molecules in a given portion of gas
at a given temperature differ only slightly from a mean speed, denoted
by ft. The ordinates of the curve in Fig. 145 represent the fractions
of the molecules which have speeds represented by the abscissae.

s 2

0-5 -

FIG. 145.— Distribution of Molecular Speeds
in a Gas.

260 INORGANIC CHEMISTRY CHAP.

It will be seen that the numbers of molecules very rapidly become
smaller which have speeds deviating appreciably from the mean
speed marked off by the vertical ordinate. If we follow any molecules
along their zigzag paths, we shall therefore find that they all
describe these with an almost constant speed, fl. The component

velocities are, of course, fluctu-
ating repeatedly, as the mole-
cule undergoes collisions, but
the speed along the path of
motion is nearly uniform the
whole time.

Calculation of the pressure of
a gas. — Consider a mass M
grams of a gas, say oxygen,
enclosed in a cube of volume
v c.c. Let there be n molecules
per c.c. of gas, the mass of each
being m gm.

Consider a square centimetre

of the surface of one of the walls of the cube and suppose
that there are n± molecules per c.c. with a component
velocity ul at right angles to this wall, n2 per c.c. with a
component velocity u2, and so on. Some of these molecules
will strike the wall, and rebound from it, and the surface will be
subjected to bombardment due to the
molecular shower.

Some molecules, however, will be
moving away from the wall, and since
there is no accumulation of gas in the
vicinity of any wall, it follows that,
on the average, exactly half the
molecules must be moving towards
the wall, and the other half away
from it. Thus, the number of mole-
cules of the first type, i.e., with
component velocities ult which take
part in the molecular shower is Jwl5
and similarly for all the other types.
If we imagine a rectangular box
erected on the square centimetre of
area as a base, and having a height ul (Fig. 146), then all the
molecules in this box which are moving towards the wall with
velocity u1 will reach the wall, and be reflected from it, in one second.
The reflected molecules move back into the bulk of the gas, and
since they will have to pass over a distance greater than ult and
back again after reflection at the opposite wall, they will not partici-

FIG. 146. — Calculation of the

Pressure of a Gas from the Kinetic

Theory.

xv THE MOTION OF MOLECULES 261

pate again in the molecular shower during the given second of time.
The last molecules of the first type to strike the wall will be those at
the extreme end of the box, distant uv because these can just- reach
the wall in one second. Molecules farther away will not reach the
wall, and this class obviously includes all molecules which have
entered the box to replace those leaving it in the opposite direction.
We need, therefore, take into account only those molecules present
in the box at the beginning of the second of time.

The number of molecules of type u^ in the box = (vol. of box)
X (No. of mols. per c.c.) =mu1nlt and the number of molecules of
the first type which participate in the molecular shower is therefore
^ulnl. Each molecule approaches the wall with velocity ul and
leaves it, after collision, with velocity — u^. The momentum
before impact is mult that after impact is — mult hence the change
of momentum on impact is mul — ( — mu-^) = 2mul. This is
balanced by an equal reaction on the wall, directed outwards from
the vessel. The total momentum given up to the wall per second
by molecules of the first type is equal to (No. of molecules of type 1
striking wall per sec.) x (momentum imparted by each molecule) =
^uln1 x 2 mUi = mn^u-f. In the same way, by considering boxes
of lengths u2, u3, etc., we can find the total momentum imparted
to the wall per second by all the molecules. This will obviously be
mfajuf _4- n2u22 + ...), which we can write in the form mnu*,
where uz is the mean square of the velocities ult uz . . ., viz.,
-2 = njuf + ttaV + • -where n + n + m m is equal to n the

% + n2 + . .

total number of molecules per c.c. The sum of the momenta
imparted per second is, however, the pressure, p, exerted by the
gas on the wall, and since mn — mass of gas per c.c. = density D,
it follows that p = Du*.

The sum of the mean square velocities at right angles parallel to
the edges of the cube is_defined as the square of the mean square
speed : 6r2 = w-2 + v2 -f- w*. On the grounds of symmetry, we may
suppose that ~u2 = v* =~w2, hence u? ' = J(72.

Thus the pressure is equal to %DG2, or to J— G2,

This is the fundamental equation of the kinetic theory of gases.

We see that the mass of gas striking the sq. cm. of the wall per second
= IfftCnjti] -j- n2u2 + ..) = \rnnu = \Du, where u is the mean of the
velocities normal to the wall, in the molecular shower. It can be shown
that uz is not equal to (u)z, but that

Vuz = 0-921 Vu*.

262 INORGANIC CHEMISTRY CHAP.

Molecular energy. — The kinetic energy of a molecule is |-ra6r2,
hence the equation we have just deduced shows that : the product
of the pressure and volume of a gas is always equal to two-thirds
of the kinetic energy of translation of the molecules. By kinetic
energy of translation we mean the energy possessed by the mole-
cules in virtue of their translatory motion in straight lines
through the gas ; only this part of their energy makes any
contribution to the pressure resulting from the molecular bom-
bardment. Energy due to the rotation of the molecules, or the
relative motions of their parts (p. 598), is without influence on
the pressure.

But from Boyle's law, pv = const, when the temperature is
constant, hence the kinetic energy of translation depends only on the
temperature of the gas, not on its volume. This is equivalent to
Joule's law, from which we started (p. 258).

Now put v = 22242 c.c., then at S.T.P., M = M, the gram-molecule
of the gas, and n = N, the number of molecules in a gram-molecule.
Avogadro's hypothesis shows that the number N is the same for
all gases ; it is called Avogadro's number, or Avogadro's constant.
We see that the kinetic energy of translation of the molecules is the
same for agm. mol. of any gas at a given temperature. For, kinetic
energy = \MG2 = f- pv. But v is the same for a gm. mol. of any
gas at a given pressure and temperature, and, by Boyle's law, pv is
also constant at a given temperature. We can now calculate this
molecular energy.

At the melting point of ice, v = 22242 c.c., p = 760 mm. = 76 X
13-59 X 980 = 1013130 dynes per sq. cm.

/. |pv = 22242 X 1013130 x f = 3-38 X 1010 ergs.

Thus, the molecular energy of a gram-molecule of any gas at 0°, due
to the translatory motion of its molecules, is large enough to raise
a weight of about one-third of a ton through one metre.

Molecular speeds. — From the value of the molecular kinetic
energy, JM6r2, which is the same for all gases, and equal (very
approximately) to 34 x 109 ergs at 0°, we can now calculate the
squares of the molecular speeds, G2, by division by the molecular

weight in grams, M, and multiplication by 2 : G2 = x 2.

Thus, in the case of oxygen, M = 32 /. G* = g4 X 109_><J

.*. the mean square speed G = 46,000 cm. per sec., or 460 m. per
sec. The mean speed, Q, is G multiplied by 0-921, i.e., 425 m.
per sec. In the case of hydrogen, the mean speed at 0° is 1700 m.
per sec.

xv THE MOTION OF MOLECULES 263

TABLE OF MOLECULAR SPEEDS AT 0° IN METRES PER SECOND.

Hydrogen, 1700 (1286) Oxygen, 425 (317)

Helium, 1213 Carbon dioxide, 362 (257)

Steam, 566-6 (400) Chlorine, 288

Nitrogen, 455 Mercury vapour, 170

It will be observed that the speed of steam molecules (M = 18) is
considerably greater than that of oxygen molecules (M = 32) ; the
speeds of hydrogen and helium are very large relatively to those of
the other gases, which may explain the small amounts of the former
gases present in the atmosphere, since these gases may have diffused
into space. A speed of 1700m. per sec. is 5500 ft. per sec., or more than
a mile per sec., i.e., of the order of speed of a rifle bullet. Owing to
these high speeds the kinetic energies of the minute fragments of matter
which the molecules represent are high, and the pressures due to the
molecular shower are thus explained. It is also seen that the molecular
speeds are of the same order as, but about 1*3 times greater than, the
velocities of sound in the gases, given in brackets after the molecular
is.

Effusion. — The relation AM6r2 = const, shows that the molecular
speeds are inversely proportional to the square roots of the molecular
weights : G1 : G2 : ' V^2 : V^r Suppose that the sq. cm. of
surface of the wall considered on p. 260 to be a little trap-door,
which is opened with a vacuum on the other side. The
molecular shower streams into the vacuum with a speed
equal to_the mean molecular velocity normal to the wall : u =

0-921/s/w2, i.e., inversely proportional to the square root of the
molecular weight. This is the phenomenon of effusion, studied by
Graham. It is not necessary that the gas shall stream into a
vacuum ; if it is forced by pressure through a small aperture in a
plate exposed to the air, the actual speed of effusion is slowed down
by collisions between gas molecules and air molecules, but
the relative rates of effusion of different gases into air are still in the
inverse proportion of the square roots of the molecular weights.
By means of this result, it is possible to compare the molecular
weights of different gases. The apparatus used, devised by Bunsen,
is called an effusiometer.

EXPT. 112. — A glass cylinder has two marks, ml9 m2, scratched upon
it, and is placed in a cylinder of water (Fig. 147). At the top of the tube
is a stopcock, communicating with the free air through a tube closed
by a thin platinum plate, in which a hole has been pierced with a fine
needle. The tube is filled with gas to a level below the lower mark, mlt
and the tap is opened. The gas streams out through the fine hole, and
the time required for the liquid surface to pass from ml to mz is taken by

264 INORGANIC CHEMISTRY CHAP.

a stop-watch. The experiment is repeated with a gas of known
molecular weight, e.g., oxygen. The ratio of the squares of the times is

the ratio of the molecular weights. If mer-
cury is used, a float is fitted inside the
tube, having a line marked on its upper end.
The time taken for this mark to pass between
two marks on the upper surface of the
cylinder is noted.

Absolute temperature. — The product pv
for a given weight of gas is proportional
to the absolute temperature : pv = R T.
But pv is proportional to the translations!
kinetic energy of the gas molecules, hence
the latter also is proportional to the absolute
temperature. Since, at constant volume,
the pressure increases by 1/273 of its
value at 0°C. for 1° rise in temperature,
the translational kinetic energy of the
molecules must increase by the same
fraction of its value at 0°C. In this
way we can easily calculate the mole-
cular speeds at any temperature from
their values at 0°C. given in the table
above.

FlQ. 147.— Bunsen's Effusio-
meter (Ostwald).

Thus, the speed of hydrogen molecules
at 1000° is found as follows : kinetic energy
at 1000° = 1273 X K.E. at 0°C. But the speed is proportional to

speed at 1000°: speed at 0

\/2T3

.'. speed at 1000° = 1700 x

= 1700 X 2-1/6 m. per sec.

The increase of speed with temperature is therefore not very rapid ;
it is doubled by a rise of 1000°.

For a gram molecule, pv = RT. The kinetic energy of translation
of the molecules is JM6r2 = %pv = ^RT. The value of R in absolute
units (p. 149) is 8-25 x 107 ergs per 1°, hence the kinetic energy
at T° absolute is f X 8-25 x 107 T ergs = 124 x 107 T ergs.
In gram calories, it is f X 1-97 T = 2-95 T gm. cal. =CVT for a
monatomic gas (p. 598).

The molecular diameter. — In spite of the high values of the
molecular speeds the diffusion of one gas into another takes place
glowly.

THE MOTION OF MOLECULES

265

EXPT. 113. — A small glass bulb containing bromine is placed in a tall
stoppered glass cylinder (Fig. 148). The bulb is broken with a glass rod,
and a layer of bromine vapour, of a dark red colour, forms at the bottom
of the jar. This vapour diffuses only very slowly upwards, although
at 17° the speed of the bromine molecules must be

1700 X

300\2 /y

tyTv ) * A / ^7: = 230 metres per sec.

&ldj /y/ 5(J

The actual rate of motion of the bromine vapour is not more than a
millimetre per second, or only about one -hundred- thousandth of the
molecular speed.

The reason for this difference is, however, clear. The molecules of
bromine do not move uninterruptedly in straight lines for indefinite
distances ; they collide with one another and with
the molecules of the air, and a great number of them
must be deflected back again to the region from which
they started. The molecules describe zigzag paths,
and it is only after making a great number of collisions
that a molecule can get appreciably forward.

The same effect is familiar to us when we walk rapidly
into a crowd of people, and if we were thrown back
every time we happened to encounter anyone else our
progress would be still further impeded.

It is clear that this effect is due to the finite size of
the molecules ; if they were mere points, occupying no
space, they would not offer any obstacles to the motions
of other molecules. It also appears probable that
from the rate of diffusion one should be able to
calculate the diameters of molecules. Clausius in this way
found that the diameter of the oxygen molecule, assumed spherical,
is of the order of 10~8 cm.

TABLE OF MOLECULAR DIAMETERS IN CM. x 10~8.

Hydrogen 2 -40 Chlorine 4-96

Helium 2-18 Carbon dioxide 4-2

Oxygen 3-4 Nitric oxide 3-4

Nitrogen 3-5 Steam 4-1

Platinum wires can be drawn to 10~4 cmi in diameter ; ordinary gold-
leaf is 10~5 cm. thick ; the black parts of soap-films are 6-10~7 cm. thick,
and oil-films on water may be only 10~7 cm. thick, or even less. The
latter contain only a few (less than 10) molecules in the thickness.

The distance of the nearest fixed star is reckoned in light-years, 1
light-year being the distance traversed by light (3 X 1010 cm./sec.)
in a year, or 1018 cm. It is therefore incorrect to regard the minuteness

266 INORGANIC CHEMISTRY CHAP.

of molecules as the counterpart of the vast interstellar distances.
The molecules are small, it is true — too small to be visible (when their
presence would be confusing), but their refinement has not been over-
done.

The mean free path. — The mean distance traversed by a gas mole-
cule before collision with another is called its mean free path, L.
This can be calculated from the viscosity of the gas, >;, by the for-
mula : L = 2-02 r)/ *JpD. It is therefore greater the lower the pres-
sure, as is obvious, because then the molecules are less crowded
together and their jostling is reduced. In oxygen at S.T.P., L is
very nearly 10~5 cm. ; it is double this in hydrogen.

The mean free path of the hydrogen molecule at atmospheric
pressure is equal to the thickness of the thinnest gold-leaf. At
low pressures, such as exist in the evacuated spaces of Dewar
flasks, the free path is several cm. A molecule rebounds from
opposite walls of such a flask many times without encountering
another molecule.

During one second a molecule describes as many free paths as it
makes collisions, and the sum of the paths is equal to the mean speed
O. Thus, the collision frequency, or the number of collisions per second,
= Q/L. In oxygen, this is 4-25 x 104/10~5 = 4-25 x 109. At
very low pressures the mean free path is 1 cm., but even then there
will be 105, or 100,000 collisions per second.

The area exposed by the surfaces of all the molecules, assumed
spherical, in 1 c.c. of oxygen at S.T.P., 4n7ir2, is about 7 square metres.

Molecular magnitudes. — The most important constants in the
kinetic theory are : n = the number of molecules per c.c. at
S.T.P. ; N = 22242 X n = the number of molecules in a gram-molecule.
The number N, which is the same for all gases, is called Avogadro's
Constant.

Until quite recently the value of N was known only approxi-
mately ; in some quarters the very existence of molecules was held
to be extremely doubtful. Within the last ten years, however, the
value of N, and hence the absolute mass of a single molecule, have
been determined by a variety of methods with an accuracy of about
1 per cent.

The most direct method used in the determination of N is
due to Rutherford and Geiger. The element radium has the
property of firing out atoms of helium with extremely high speeds.
These atoms, called a-rays, move with speeds of about 2 x 109
cm. per sec. (i.e., about 100,000 times faster than gas molecules),

THE MOTION OF MOLECULES

267

and their kinetic energy is therefore extremely large. If the a-rays
from a particle of radium, A, are allowed to impinge on a screen of
zinc-blende, B, in the spinthariscope of Crookes (Fig. 149), each
particle causes a flash of light easily visible under a lens, C. It
was therefore possible to count the a-rays emitted in a given time,
and by collecting the helium from a large amount of radium over
a long period, the volume of helium
produced from 1 gm. of radium
was found to be 046 cu. mm. per
twenty-four hours. By comparing
this with the directly counted num-
ber of a-particles (helium atoms) FIG. 149.— Spinthariscope.
emitted from a known weight of

radium in a given time, it was easy to calculate the number of
molecules (atoms) per c.c. of helium. This is n, and its value
came out at 2'7 X 1019. Thence N = 6 05 X 1023.

A second method used by Rutherford and Geiger (1908)
depends on the capacity of the rapidly moving a-particles of rendering
a gas through which they pass a conductor of electricity (p. 1021).
A long glass tube, AA/ (450 cm. long and 2-5 cm. wide), called the
" firing tube " (Fig. 150), was exhausted, and at the end A was
placed a preparation of radium on a lead plate, a, which expelled
a-particles. Some of these were shot along the tube and passed

450 cm.

FIG. 150.— Rutherford and Geiger's Apparatus.

through the narrow tube, B, into the brass ionisation chamber
C, where the gas at low pressure was rendered conducting, or was
ionised. A mica window at .F shut off the gas from the evacuated
tube, ^4^4'. Running axially through the vessel (7, and insulated
from it by the ebonite ends, was a metal wire, w, which was connected
through a battery and electrometer to the outer surface of the brass
vessel. As each a-particle entered the ionisation chamber (at the
rate of about one every second), it made the gas conducting, and the
electrometer gave a deflection. In this way the individual a-rays

268 INORGANIC CHEMISTRY CHAP.

were counted, and the method of calculation was similar to that in
the first method. The value N = 6-09 X 1023 was found.

The determinations of N have been made by counting, as above,
and from other radioactive experiments, from experiments on colloidal
solutions (p. 311), the spectrum, the radiation of heat, the formation
of clouds, and the blue colour of the sky. The numbers obtained
from the recent experiments are in excellent agreement, and leave
no doubt that the latter cannot possibly be the result of chance.
Everything points to the real existence of molecules. Avogadro's
hypothesis may now be regarded as a law, and an undue insistence on
the hypothetical character of the atomic and molecular conceptions
of the structure of matter is belated, and out of touch with modern
experimental science. The diversity of methods by which N has
been found, only a few of which are referred to above, illustrates the
fundamental character of the molecular theory in all branches of
physics and chemistry.

TABLE OF VALUES OF AVOGADRO'S CONSTANT, N.

METHOD. N.

Classical Kinetic Theory 10 X 1023 (approximately)

Cloud Formation (p. 1024) 8-3 X 1023

Brownian movement (p. 311) 6-25 X 1023

Radiant heat 6-14 X 1023

Counting a-particles 6-09 X 1023

Electronic charge (Millikan, p. 281) 6-03 X 1023
The most recent measurements agree to within 1 or 2 per cent.

TABLE OF MOLECULAR MAGNITUDES.

Number of molecules per c.c. of gas at S.T.P. = n == 2-70 x 1019.
Number of molecules per gram-molecule (22-24 litres in ideal state

at S.T.P.) = N = 6-03 "x 1023.
Mass of hydrogen atom = 0-000089873/(2 X 2-7 x 1019) =1-67 x

10-* gin.

Mean speed of hydrogen molecule at 0° = QH2 = 16-94 x 104 cm. /sec.
Translational kinetic energy of a molecule at 0° = E0 = 33-85 X

109/6-03 x 1023 = 5-613 X 10~14 ergs (p. 262).
Rate of change of translational kinetic energy per 1°= e =

5-613 X 10~14/273-09 = 2-056 X 10~lg ,erg

degree

A few special magnitudes, not known with the accuracy of the above,
may be given for comparative purposes :—

xv THE MOTION OF MOLECULES 269

Diameter of hydrogen molecule =2-17 X 10~8 cm.
Mean free path of hydrogen molecules at S.T.P. = 1-42 X 10~5cm.
Average distance apart of gas molecules at S.T.P. = 3 X 10~7 cm
Number of collisions per second of hydrogen molecules at S.T.P. =

1-2 X 1011.
Time of describing free path -of hydrogen molecules at S.T.P. =

3 X 10-10 sec.

Molecular attraction.— We have assumed so far that the forces
exerted by gas molecules on one another are negligibly small. This
is only approximately true. Gases are usually more compressible
than according to Boyle's law, and this indicates that the molecules
attract one another. This attraction becomes greater the closer the
molecules come together ; when the gas is liquefied the molecular
attraction is sufficient to prevent the molecules flying off into space,
as they do in an open vessel of gas. But a liquid is very much less
compressible than a gas, and the compressibility of a gas falls off
considerably at high pressures (p. 66). This effect is assumed to be
due to the space occupied by the molecules, x ; if this is comparable
with the total space, v, we shall have only the intermolecular space
(v — x) available for compression.

These two factors are taken into account by the equation of Van
fler Waals. In this, the ideal gas equation pv = RT is replaced by

where a and b are constants : a/v2 is the molecular attraction cor-
rection, which is inversely proportional to the square of the volume ;
it adds itself to the external pressure : b is the correction for the
space occupied by the molecules ; according to Van der Waals, b is
equal to four times the total volume of the molecules, but it
appears to be 4\/2 times the latter. This equation gives very good
results with some gases (e.g., ethylene), but there is no doubt that
the attraction term depends on the temperature. D. Berthelot
has used the equation :

+ 35-,) <« - b) = RT,

with remarkably good results at moderate pressures.

Changes of state. — The attractive forces exerted by molecules
upon one another are of considerable magnitude, on account
of their propinquity when the substance is in the liquid or solid
state. In a liquid we may suppose that the molecules are lying
close together, so that there are practically no free paths. The
motion of the molecules is now more analogous to gliding of the
particles among and over one another.

270 INORGANIC CHEMISTRY CHAP.

In the liquid state the molecules exert attractive forces on each
other, but a molecule in the body of the liquid is attracted equally
in all directions, and the resultant force on it is zero. The range
of these attractive forces is small ; Van der Waals has calculated it
to be of the order of 10~6 cm. Those molecules lying in the surface of
the liquid, however, are subjected to a resultant attraction, due to
the unbalanced forces of the molecules beneath them, and are under
a pressure tending inwards towards the body of the liquid (Fig. 151).
It is this resultant force which, as is explained in text-books on
physics, gives rise to the phenomena of surface tension.

Recent investigations appear to lead to the conclusion that the attrac-
tive forces between molecules are not exerted uniformly in all directions,
but proceed along rays in one or two directions only, as if .the molecules
were small magnets. The molecules in the surface will then mostly
be arranged with the same parts pointing in one direction.

Some of the molecules in the liquid will possess more kinetic
energy than the rest, although most of them possess kinetic energies

close to a mean value. It may
happen that such a molecule, ap-
proaching the surface, will have
sufficient energy to break away from
the attractive forces, and it will
proceed outwards into the space
above the liquid. This is the phe-
nomenon of evaporation.

_____ n^catin7the~ , . This escaPe of_ molecules of higher

Range of Molecular Forces in a Liquid, kinetic energy than the average will

obviously reduce the mean energy

of the liquid, which becomes cooler. To maintain the temperature
constant, heat must be added from outside ; this is the latent heat
of evaporation.

Molecules in the vapour approaching the liquid will be attracted
when they come near the surface. They will then describe curved
orbits, and in many cases will be caught by the surface and dragged
into the liquid. They experience an acceleration in the field of
attraction, and pass into the liquid with increased kinetic energy.
Heat is therefore given out on condensation. Eventually, a state is
reached when as many molecules leap out of the liquid as are dragged
back again per second ; this is a condition of equilibrium, corre-
sponding with the maximum, or saturation, vapour pressure, but it is
a kinetic equilibrium, due to two opposite processes, evaporation and
condensation, going on simultaneously to equal extents.

xv THE MOTION OF MOLECULES 271

In the solid state it is assumed that the molecules are fixed in
definite positions, each molecule performing oscillations of small
amplitude about its position of equilibrium. When heat is imparted
to the solid, the amplitude of these oscillations increases, and at a
certain temperature the molecules perform oscillations of such ampli-
tude that they collide with each other, and begin to break loose.
This is the point of fusion at which the solid passes into the liquid
state, when the molecules glide about amongst each other. The
process of solidification consists in the liquid molecules building
themselves up again into a system of molecules oscillating about fixed
points. The solid molecules exert considerable attractive forces
upon each other ; in separating them under the influence of these
forces work is done, which is equivalent to the latent heat of fusion.

The process of rebuilding the solid structure from the liquid takes
place around definite points or nuclei. Small crystal aggregates are
formed at a comparatively small number of points, and radiating masses
of crystals shoot out from these centres until the whole mass is solid.
This process can be examined under the microscope, and the appear-
ance is very striking and beautiful. Crystallisation does not usually
begin at the freezing point unless solid is present ; the liquid must be
supercooled before solid appears. A solid, on the other hand, always
fuses as soon as the melting point is reached, and cannot be permanently
superheated. The temperature of the supercooled liquid rises to the
melting point when 1 he first portion of solid appears.

At the melting point, when both solid and liquid are present, there
is a condition of kinetic equilibrium similar to that described in
connection with a liquid in contact with its vapour.

Solution. — When a gas is brought in contact with a liquid, solution
occurs until the concentration of gas dissolved in the liquid is in a
constant ratio to that in the gas-space, as required by Henry's
law (p. 96). A state of equilibrium is set up : Gas ^± Gas (dissd.),
but there is no reason to doubt that this is a kinetic equilibrium, as in
the case of evaporation, the same number of gas molecules entering
and leaving the liquid through the surface of separation in unit time.

The mass of gas impinging on the liquid surface per second is \Du
(p. 261) = iZ>(0-921 \/^) =iD(0-921 VW2) = 0-266 DG. In the case
of oxygen at S.T.P., D = 0-001429 gm. per c.c., G =4-61 X 10* cm.
per sec., .'. the mass of oxygen striking 1 sq. cm. of the liquid surface
per second is 0-266 X 0-001429 X 4-61 X 104 gm. = 17*5 gm. This

will container- 24 = 3-3 x 1023 molecules, or the number

oZ X 1'DO X 1U
of molecules in about 12 litres.

The molecules striking the surface of the liquid may rebound to

272 INORGANIC CHEMISTRY CHAP.

a certain extent into the gas-space, but a certain proportion pass
through the surface into' the liquid, owing to the molecular attraction
between the molecules of the gas and those of the liquid.

This is the phenomenon of solution. Of the molecules of the gas
moving about in the liquid, some will be approaching the surface, and
if the kinetic energy of any one of these is above a certain value, it
will leave again and pass back into the gas-space. This will occur
the oftener the more gas molecules are dissolved. A state of kinetic
equilibrium is reached when equal numbers of molecules leap into
and out of the liquid per second.

Now let the pressure of the gas be raised. The number of mole-
cules per c.c., or the concentration, is increased, and the number
striking the surface becomes larger in the same ratio, since it is
proportional to Z). The number of molecules per c.c. in the liquid
is also increased. By reason of this, more molecules leave the liquid
than previously. When equilibrium is established, the same number
leave as enter, per second, but if the number entering had been
increased n times the number per c.c. of liquid will have been in-
creased n times. This is Henry's law.

At first sight it may seem that the gas could have any concentration
in the liquid, since as many molecules enter as leave. But if we
imagine people walking into a room through one door and out through
another, so that as many enter as leave, then if they enter twice as fast
there will be double the number in the room, although they are also
leaving it at twice the previous rate.

The solution of a solid in a liquid may be considered from the
same point of view. Molecules are torn away from their centres of
oscillation on the surface of the solid, and molecules are caught into
positions of oscillation out of the liquid. In this case the kinetic
nature of the equilibrium in a saturated solution can be observed,
because if an irregular or broken crystal is suspended in a saturated
solution, it tends to become more perfect in shape, one portion
dissolving and being deposited again in another place.

As to the effect of temperature on solubility, the kinetic theory
in its present stage gives no useful information, and we shall omit
further description of this subject.

SUMMARY OF CHAPTER XV

The molecules of all bodies, at temperatures above the absolute zero,
are in motion. Those of a gas exert practically no forces on each other
unless the gas is strongly compressed, and the pressure exerted by a
gas is due to the bombardment of the walls of the containing vessel by
the molecules.

CH. xv THE MOTION OF MOLECULES 273

If p is the pressure, D the density, of the gas, the mean square speed,
G, of the molecules at any given temperature can be calculated from
the equation : p = ^DG2. The mean speed, &, of the molecules is
0-921V(52. At 0° the speed of the hydrogen molecule is 1700 m. per
sec. ; those of other molecules are inversely proportional to the square
roots of the molecular weights.

The kinetic energy of translation of the molecules in 1 gm. mol. of
gas is plCr2, where M = mol. wt. ; this depends only on the tempera-
ture and is the same for all gases. At 0° it is 3-38 X 1010 ergs.

The velocities of effusion of two gases are inversely proportional to the
square roots of the molecular weights.

The molecular diameter is of the order of 10~8 cm. ; the mean free
path, i.e., the distance traversed by a molecule before collision, is about
10~5 cm. at S.T.P.

Avogadro's constant, N, is the number of molecules in a gm. mol. ;
with a probable accuracy of 1 per cent, it is 6-03 X 1023.

The molecules of liquids and solids are much closer together than those
of gases, and exert attractive forces on one another.

EXERCISES ON CHAPTER XV

1. What evidence is there that the molecules of gases and liquids are
in motion ? What is the speed of hydrogen molecules at 0° ? How
do you explain the fact that hydrogen diffuses through air at a much
slower rate than this ?

2. How is the pressure of a gas accounted for on the kinetic theory ?
Show how the pressure may be calculated from the molecular velocity.

3. What relation is there between the pressure of a gas and the
kinetic energy of its molecules ? How is the temperature of a gas
represented on the kinetic theory ?

4. By what methods has the molecular diameter been determined ?
What is its approximate value, and how near has this been approached
in actual bodies ?

5. What is Avogadro's constant ? How has it been determined ?

6. What evidence is there for the existence of molecular attraction
(a) in gases, (6) in liquids ? How does the kinetic theory explain
evaporation and crystallisation ?

7. Show how the effect of pressure on the solubility of a gas may be
deduced from the kinetic theory. How would you explain the devia-
tions from the law ?

8. 17-91 c.c. of chlorine were mixed with a given volume of oxygen,
and allowed to diffuse into a vessel of oxygen for forty-five minutes.
4-05 c.c. of chlorine diffused in this time into the second vessel.
The same experiment was carried out with 22-57 c.c. of carbon dioxide,
and 6-67 c.c. were found to have diffused in forty-five minutes. Find
the ratio of the molecular weights of chlorine and carbon dioxide.

9. What is effusion ? How may the molecular weights of gases be
compared by their relative rates of effusion ?

CHAPTER XVI

ELECTROLYSIS

The dualistic theory of Berzelius. — Lavoisier showed that non-
metals (except hydrogen), when burnt in oxygen, yielded acidic
oxides which produced acids with water. He regarded oxygen
as the principle of acidity (Greek oxus, sour). Davy found that
sodium and potassium burnt in oxygen to form basic oxides, which
gave alkalies with water, hence oxygen is also a constituent of bases.
When baryta, or barium oxide, which is a basic oxide, is mixed with
the acidic sulphur trioxide, both being solids, the mass becomes
red-hot, and the neutral salt, barium sulphate, is formed : BaO +
S03 = BaS04. Lavoisier considered salts as compounds of acidic
and basic oxides, e.g., BaO,SO3, and this idea of
two parts contained in a salt was amplified by
Berzelius (1811) into what was called the dualistic
system.

Berzelius found that solutions of the salts of the
alkalies, when decomposed by an electric current,
liberated alkali at the negative pole, and acid at
the positive pole, and he therefore considered that
the alkali and acid possessed positive and negative
charges, respectively, and that these were drawn to
the poles by the attraction of unlike charges

FIG. 152— U -tube
with Electrodes.

EXPT. 114. — Pour a solution of sodium sulphate,
coloured purple with neutral litmus, into a U-tube
with electrodes (Fig. 152), and connect with the
terminals of a battery, or the lighting mains. Observe that the liquid
around one (the positive) pole becomes red, showing that an acid
(sulphuric acid) is set free, whilst that surrounding the other (negative)
pole becomes blue, from liberation of alkali (caustic soda). Notice also
that oxygen and hydrogen are liberated at these poles, respectively.

When a metal was deposited, as from copper sulphate, it was sup-
posed to have been formed from the oxide, CuO, by reduction with
the hydrogen, which in such cases is not evolved. The hydrogen
and oxygen, it was thought, came from the water.

274

CH. xvi ELECTROLYSIS 275

According to the theory of electrochemical dualism, salts are binary
compounds of two oxides, the acid and the base, which are them-
selves binary compounds of elements with oxygen :

Sulphate of soda Na2O,SO3.

+ +

Soda Na2 -f O. Sulphuric acid S + O8.

Elements giving basic oxides were called electropositive elements,
those giving acidic oxides were called electronegative elements.
Oxygen was assumed to be always electronegative ; it was " the
pole around which the whole chemical system revolved." The

radation of electrochemical character was expressed in the table

)f elements given on p. 133.

This dualistic system was soon shown to be untenable in its
original form. Its downfall was brought about by three circum-
stances : (1) the recognition of the elementary nature of chlorine,
which, since it forms salts, had previously to be regarded as an acidic
oxide of an unknown element ; (2) the discovery of the true character
of electrolysis, which accounted for the simultaneous production
of hydrogen and oxygen in the decomposition of salts ; (3) the
investigation of substitution reactions in organic chemistry — thus,
an electronegative atom of chlorine can replace an electropositive
atom of hydrogen without altering very much the chemical nature of
the compound (p. 398).

C2H40? + C12 = C2H3C102 + HC1.
Acetic acid Chloroacetic acid

Many complicated equations involving oxidations and reductions
are, however, most simply written down by making use of the
obsolete dualistic notation, and the latter is still of service in this
way (p. 969).

The electrolysis of sodium chloride solution. — According to Berze-
lius's dualistic theory, it might be expected that the electrolysis of a
solution of sodium chloride would yield caustic soda and hydrochloric
acid. The reaction, however, is different.

EXPT. 115. — Repeat Expt. 114, with a solution of common salt in
the U-tube. The litmus around the negative pole is turned blue, from
liberation of caustic soda, but that around the positive pole is bleached,
indicating that chlorine is evolved. Hydrogen is evolved from the
negative pole.

It appears that the salt is decomposed with liberation of chlorine,
and the sodium first set free at the negative pole then reacts
with the water to give caustic soda and hydrogen, which are
actually liberated at that pole. The primary production of sodium
at the negative pole can, in fact, be demonstrated.

T 2

276

INORGANIC CHEMISTRY

CHAP.

EXP T. 1 1 6. — Pour a little mercury into a glass tube having a platinum
wire sealed through the bottom (Fig. 153). Fill up the tube with sodium

chloride solution, and connect the wire
with the negative pole of a battery of
two accumulators. The positive pole
is connected with a piece of platinum
foil dipping into the solution. The
liquid soon smells strongly of chlorine,
but very little gas is evolved from the
mercury. The sodium liberated dis-
solves in the mercury and forms an
amalgam. After a few minutes stop
the experiment, and pour the mercury
into water? Bubbles of hydrogen are
evolved, and the water turns red
litmus blue, showing that sodium was
present, which reacts with the water.

Fla 1B3^uryr°cftLSoSbe with The sodium chloride in EXPT. 116

is decomposed by the current into

sodium and chlorine, which are deposited at the negative and
positive poles respectively. The atoms of chlorine combine to form
molecules of chlorine gas, which is evolved. The atoms of sodium
at once react with the water present, forming caustic soda and
liberating hydrogen, which is evolved :

2NaCl
* \
Neg. pole H2 + 2NaOH <- 2H2O + 2Na 2C1 -> Cl2gas Pos. pole

The primary products of the electrolysis are sodium and chlorine ;
the sodium reacts with the water to give hydrogen and caustic soda,
which are secondary products.

J. F. Daniell, of King's College, London, suggested in 1840 that
the decomposition of %all salts proceeds on the same lines as that of
sodium chloride, and that the acid and the base, regarded as primary
products by Berzelius, were really secondary products. Sodium
sulphate he regarded as a compound of sodium and the radical S04,
instead of a compound of soda, Na2O, and sulphuric anhydride, SO^,
so that its formula is Na2-S04. This is decomposed by the current,
primarily, into its two radicals, which then react with water to form
soda, sulphuric acid, and the two gases hydrogen and oxygen :

H2 + 2NaOH <- 2H 20
Neg. pole

Na2S04
^ ^

2Na S04 + H20 -> H2S04 + 0
Pos. pole

XVI

ELECTROLYSIS

277

All salts were therefore regarded as constituted on the same plan
as common salt, whereas the latter was regarded by Berzelius as an
exceptional type. This theory was extended by Daniell to the acids ;
the latter were regarded as salts of hydrogen. DanielFs theory was
shown to be correct, and the dualistic theory of Berzelius was given
up.

Electrolysis. — The fundamental laws of electrolysis were dis-

MICHAEL FARADAY.

covered by Michael Faraday, whose results were published in 1832.
He introduced a number of new names, which are still used in
describing the phenomena, and must be mentioned before his
conclusions are stated.

Conductors of electricity are of two kinds : (1) those which conduct
the current without undergoing chemical change, and are simply
heated by the passage of the current ; metals and graphite belong to
this class, the members of which are called metallic conductors :

278

INORGANIC CHEMISTRY

CHAP.

(2) those which are decomposed by the current, such as acidulated
water, and solutions of salts. Conductors of the second type
Faraday called electrolytes (Greek lysis, setting free). This name is
now used to denote the dissolved substances themselves ; thus, com-
mon salt and sulphuric acid are called electrolytes, because when

dissolved in water they form
electrolytically conducting solu-
tions. In electrolysis one
portion of the products ot de-
composition travels to the
positive pole, or positive
electrode (Greek hodos, an en-
trance), or anode (Greek ana,
up) ; the other portion travels
to the negative electrode, or
cathode (Greek kata, down).
These atoms, or groups of
atoms, which wander through
the electrolyte Faraday called
the ions (Greek ion, a wanderer) ;
the anions are those which ap-
pear at the anode, and the

those which appear at the cathode. A diagram
this nomenclature is shown in Fig. 154. No

ELECTRODES

- - i~>rx"-

> ELECTROLYTE

CATION (J) 5»~

-< QANION

ANODE (+) CATHODE(-)

. 154. — Nomenclature of Electrolysis.

Sn CL

cations are

illustrating

chemical action is perceptible in the body of the electrolyte,

but only at the electrodes.

Faraday connected in
series a number of electro-
lytic cells, containing dif-
ferent electrolytes, with a
battery and an ammeter
for measuring the current
by its magnetic action,
as shown in Fig. 155.
Suppose, for instance, that
the first cell contains water
acidulated with sulphuric
acid, the second a solution
of copper sulphate, and
the third fused stannous
chloride. Fused salts are electrolytes, as well as their solutions.

After the current has passed for a certain time, the products of
electrolysis, which are liberated at the electrodes, can be measured.
Thus, the volumes of hydrogen and oxygen liberated from the
acidulated water, and the weights of copper and tin deposited from
the solution of copper sulphate and the fused stannous chloride,

FIG. 155. — Diagram of Electrolytic Circuit.

xvi ELECTROLYSIS 279

respectively, can be ascertained. The quantity of electricity which
has passed through the solution is measured by the current strength
multiplied by the time. The current strength is measured in
amperes, and one ampere passing for one second corresponds with
unit quantity of electricity, or one coulomb. A current of C
amperes flowing for t seconds conveys Ct coulombs.

If the current passes until 1 gm. of hydrogen has been liberated
from the acidulated water it will be found that 96,000 coulombs of
electricity have passed through the cells. Thus. 96,000 coulombs
liberate 1 gm. of hydrogen. If this quantity of electricity passes
as a small current for a long time (e.g., 0-1 ampere for 960,000 sees.)
or as a large current for a shorter time (e.g., 10 amperes for 9,600
sees.), the result is the same. Hence the weight of an ion deposited
in a given time is proportional to the strength of the current. This is
Faraday's First Law of Electrolysis.

If the weights of the other ions which are deposited in the cells
whilst 1 gm. of hydrogen is liberated in the first are determined, it is
found that they are equivalent weights: 7-94 gm. of oxygen,
35-2 gm. of chlorine, 31 -5 gm. of copper, and 58-9 gm. of tin. Thus :
96,000 coulombs liberate one gram-equivalent of any ion in electrolysis.
This is Faraday's Second Law of Electrolysis.

The quantity of electricity 96,000 coulombs is fundamental in
electrolysis, and is called a faraday, denoted by F. Thus, one F
liberates 1 gm. atom of a univalent element, and nF liberate 1 gm.
atom of an n-valent element.

EXAMPLE. — Find the weight of copper deposited from a solution of
copper sulphate by a uniform current of 0-25 amp. flowing for one hour.
Quantity of electricity passed through electrolyte = 0-25 X 60 x 60

= 900 cmb.

Copper is bivalent, hence equivalent weight = at. wt. -f- 2 = 63-1/2 =
31-5.

96,000 cmb. liberate 31-5 gm. of Cu, hence wt. of copper liberated by
900 cmb. = 31-5 X 900/96,000 = 2-95 gm.

Theory of electrolysis. — The facts of electrolysis are summarised in
the two laws of Faraday. An explanation of the phenomena must
include these laws. Since the ions are attracted by tjie electrodes, it
is simplest to assume that they are themselves charged, the sign of
the charge on an ion being opposite to that of the electrode towards
which it moves. Thus, anions are negatively charged atoms or
radicals ; cations are positively charged atoms or radicals. In the
electrolyte we must therefore picture two streams of charged ions

280

INORGANIC CHEMISTRY

CHAP.

moving in opposite directions to the two electrodes (Fig. 156).
These streams of charged ions constitute the current in the electro-
lyte ; the electricity is ferried across from one electrode to the other
by the charged ions, and this convective current completes that
passing through the metallic circuit outside the cell. The strength
of the current is uniform throughout the whole circuit, whether the
latter is all metallic, or composed of metal wires and electrolytes.
Since the current in the electrolyte is composed solely of chaVged
ions, the weight of the latter moving to the electrodes in a given
time is proportional to the current strength. This is Faraday's
First Law.

When a positively charged cation touches the cathode, its charge
passes into the latter, which is able to conduct the electricity without
simultaneous movement of ions. The negatively charged anion
touching the anode also gives up its charge, and the two uncharged
atoms or molecules are liberated at the electrodes. They may then
react with the water to form secondary
products.

Faraday's Second Law is simply explained
by the assumption that the quantity of
electricity associated with an ion is the
same for all ions of the same valency, and
is proportional to the valency. Thus, a
univalent cation such as sodium carries
one unit charge of positive electricity, a
bivalent cation such as copper carries two
unit charges of positive electricity, and
so on. A univalent anion, such as chlorine,
carries one unit charge of negative electricity,
which is equal in magnitude but opposite in sign to the charge on
the univalent cations, whilst a bivalent anion such as the sulphuric
acid radical, S04, carries two unit negative charges, and so on.

The ionic charges carry with them the matter with which they are
associated. When the ions reach the electrodes, the charges leave
them, and the matter is deposited. Since the current is uniform
throughout the circuit, the quantities of the ions deposited must all
be proportional to the amounts associated with the same quantity
of electricity. According to the theory advanced above, these
amounts are in the proportion of the chemical equivalents. Thus,
the same current deposits amounts of the ions which are proportional
to the chemical equivalents. This is Faraday's Second Law of
Electrolysis. The quantity of electricity associated with 1 gm.
equivalent of an ion is found experimentally to be 96,000 coulombs.

•f

f
+

0* ® *e

+
4-

*G ©•* 0*

+
[+

FIG. 156.— Migration of Ions
in Electrolytic Cell.

The ionic charges are large. To liberate 1 gm. of hydrogen, the
current which lights an electric lamp (0-5 amp.) would have to pass for

xvi ELECTROLYSIS 281

nearly fifty-four hours. If charges equal to that associated with 1 mgm.
of hydrogen could be imparted to each of two small spheres placed
1 cm. apart, they would repel each other with a force of about 1010
tons weight. As Faraday remarks, the electric charges concerned in
the most violent flash of lightning would barely serve to decompose a
single drop of water.

Electrons. — The unvarying amount of the electric charge on
univalent ions, and the simple multiple relation between the
charges on multivalent ions, suggest at once that electricity,
like matter, is divided up into atoms. It might be supposed
that there were two kinds of unit charges, one positive and
the other negative. A cation would then be an atom or radical
plus one positive unit ; and an anion would be an atom or
radical plus one negative unit. This hypothesis of the atomic
structure of electricity originated with Helmholtz (1880) : it is
a simple outcome of Faraday's results. The view was regarded
with scepticism until J. J. Thomson, in 1895, succeeded in actually
isolating the unit, or atom, of negative electricity, which is called an
electron. This is the only kind of electricity yet isolated in the free
state ; a positive charge is always associated with matter, and a
positively charged body may thus be regarded as matter which has
lost free negative electricity. A negative ion, or anion, is then
regarded as an atom or radical plus one or more electrons ; a positive
ion, or cation, is an atom or radical which has lost one or more
electrons. For convenience the charge of an ion is represented by dots
or dashes placed over the symbol ; one dot denotes unit positive
charge, one dash unit negative charge. These symbols are given
below on the right.

If the electron is denoted by the symbol €, the constitution of
ions may be represented as follows :

chloride ion = chlorine atom -f- electron = Cl + € — Cl'
hydrogen ion = hydrogen atom — electron = H — € = H\
ferric ion = iron atom — 3 electrons = Fe — 3e = Fe'".
ferrocyanide ion = ferrocyanide radical -f~ 4 electrons =

Fe(CN)6 + 4€ = Fe(CN)6"".

It has been shown that the electron is material in the sense of
possessing a definite mass. This is very small, being only 1/1845
that of the hydrogen atom. The atomic weight of the electron is
therefore 0-00054. Its absolute mass is therefore 0-00054 X 1-66
X 10-24 = 8-9 X lO'28 gm. The radius of the electron has been
calculated as 1-9 X 10"13 cm.

The electronic charge. — Since 1 gm. of hydrogen is associated, in the
ionised condition, with 96,000 coulombs of electricity, and since this
weight of hydrogen contains N = 6-03 X 1023 atoms, it follows that

282 INORGANIC CHEMISTRY CHAP.

the value of the unit charge, in coulombs, is 96,000/6-03 X 1023 =
1-592 X 10 19 coulombs.

In calculating from the relation P = Nf, used above, the value of F
has been taken as 96,000. This is based on the International Ampere,
which depends on the deposition of silver from a salt by electrolysis.
The international ampere is defined as the current which, flowing
uniformly for one second, deposits 0-0011180 gm. of silver. The value
of the faraday, F, thus depends on the atomic weight of silver. In the
International Tables this is given as 107-88 (O = 16), i.e., 107-04 (H = !)•
Thus, the value of the faraday will be :

P(6 = 16) = 107-88/0-001118 = 96,500 coulombs ;
\ m F(H = 1) = 107-04/0-001118 = 95,770 coulombs.

The value 96,000 (H = l) is sufficiently accurate for all practical
purposes.

The value of the charge on the electron has been deter-
mined in different ways, notably by Millikan, professor of physics
at Chicago (1912), who used the following very direct method.

Two metal plates, separated by
a distance of about 1 mm., were
charged positively and negatively,
respectively, by attaching them to
the poles of a battery. Into the
air between the plates a fine dust
of pulverised oil was blown by a

FIG. 157.-Millikan's Determination of" SPray- The oil dr°PS' which Settled

the Electronic Charge. very slowly on account of their

small size, were found to be electri-
cally charged. A particular drop was focussed in the field of a
microscope with a scale in the eyepiece, as shown diagrammatically
in Fig. 157. By varying the potential difference between the plates,
the charged drop could be made to move upwards or downwards
with any desired velocity, or kept suspended. From the difference
between the velocities 'of fall, with and without the potential
difference, the charge on the drop could be calculated.

It was found that this charge was not constant, but varied during
an experiment. The important thing, however, was that these varia-
tions were not continuous, but took place in jumps. Each sudden
change was assumed to correspond with the gain or loss of one or more
electrons by the drop, and it was found that the charge varied in
small multiples of 1*59 X 10~19 coulombs. Thus, the value of the
charge on the electron is 1-59 X 10" 19 coulombs.

In the above calculation, the value of N, which is derived from that of
€ by the relation F = Nf, was that found from Millikan 's value of e. The
values of N and e can, however, be determined in other ways. The

xvi ELECTROLYSIS 283

value of e determined by Rutherford and Geiger, by counting the
a-particles emitted from radium (p. 267), was 1-55 x 10~19 cmb.

Electrolytic dissociation. — The picture of the mechanism of
electrolytic conduction employed above implies that the ions move
independently through the electrolyte. They behave as if they
were free, and each ion responds to the attraction of the electrodes
as if the other ions were not present. If the current is switched off,
no visible- change occurs in the solution, so that we may assume that
the ions still remain in the solution free and independent of each
other.

Clausius (1857) assumed that in the solution of an electrolyte a
few molecules of the salt are broken up into ions, the processes of
decomposition and recombination going on continually, and tlje
free ions present at any instant are transported as the current.
Williamson (1851) had previously assumed an exchange of atoms
between different molecules of the electrolyte, and thought that
during the exchange the atoms or radicals existed transitorily in
the free state. He assumed, however, that this exchange occurs
also in gases. It was Arrhenius, in 1887, who first made the bold
assumption that nearly all the molecules of the electrolyte may be
dissociated into free ions.

According to this theory of electrolytic dissociation, or of ionisation,
an electrolyte (salt, acid, or base), when dissolved in water or certain
other solvents which yield conducting solutions (such as ethyl and
methyl alcohols, pyridine, anhydrous hydrocyanic acid, or form-
amide), undergoes a chemical change in such a way that from the
electrically neutral molecule two or more ions are produced. The
sum of the positive and negative charges on the ions must always be
zero, since the solution as a whole is uncharged.

The current in the solution is due solely to the free ions ; the
undissociated salt molecules do not move to the electrodes. When
the ions reach the electrodes their charges are neutralised, and the
uncharged atoms or molecules are deposited. The process of
electrolysis can, therefore, be represented diagrammatically as in
Fig. 156. Thus, sodium chloride, when dissolved in water, is largely
ionised into the sodium ion and the chloride ion : NaCl = Na* + Cl'.
This takes place whether the solution is electrolysed or not. In
electrolysis, the negative chloride ions are attracted to the positive
anode, and on reaching it give up their charges, becoming chlorine
atoms : Cl' = Cl -f e. These cannot exist as such, but combine
in pairs to form chlorine molecules, which escape as chlorine gas.
The positive sodium ions, on reaching the cathode, take from it the
negative charges, or electrons, which have passed round the metallic

•

284

INORGANIC CHEMISTRY

CHAP.

circuit from the discharged chloride ions, and so become neutral
sodium atoms : Na* -f- c = Na. These may dissolve in mercury,
if the cathode is metallic mercury ; or react with water, forming
caustic soda and hydrogen, if the electrode is of platinum.

The atoms of the substances, at the moment of liberation at the
electrodes, may be very reactive. Thus, hydrogen liberated by the
electrolysis of an acid can bring about the reduction of a ferric salt
added to the solution, in the same way as nascent hydrogen (p. 189).

The extent of ionisation of a dissolved electrolyte is called the
degree of ionisation, and is denoted by a ; it corresponds with the
degree of dissociation of a gas, y. Thus a = ionised part of
electrolyte/total amount of electrolyte. A solution of potassium
chloride containing 0-001 gm. mol. per litre is ionised to such an
extent that of every 100 molecules of KC1 dissolved only 2 remain
undissociated and 98 are broken up into ions. The degree of
ionisation in this case = 0-98, or 98 per cent.

The ionisation of a dissolved electrolyte is entirely different from
the thermal dissociation of a gas. Thus, ammonium chloride on
heating dissociates into ammonia and hydrochloric acid : NH4C1 =
NH3 -j- HC1, but in solution it is electrolytically dissociated into the
ammonium and chloride ions : NH4C1 = NH4* -j- Cl'. It therefore
seems expedient to refer to the latter change as ionisation, although
this name has recently been used for a different change occurring
in gases exposed to Jf-rays or radioactive substances, and so rendered
conductors of electricity (p. 1021).

The reader will have no difficulty in representing the reactions at
the electrodes during the electrolysis of salts by means of the ionic
theory. The electrolysis of copper sulphate may be taken as an
example :

<- 2€

Cathode

CuS04

Cu
deposited

S04"

| -> 2

S04

reacts with
water :

Anode

S04

H20 = H2S0
evolved.

-|0a

The nature of the ions. — The question at once arises as to how it
is possible to have in an aqueous solution of common salt either free
sodium or free chlorine, since the former is violently attacked by

xvi ELECTROLYSIS 285

water, and the latter is a greenish-yellow gas, forming a greenish-
yellow solution with water. The solution shows none of the proper-
ties of sodium or -chlorine. The answer is that neither metallic
sodium nor chlorine gas are assumed to be present in the solution,
but only sodium ions and chloride ions. These differ from the free
elements by possessing large electric charges. It has already been
emphasised that ferrous and ferric salts behave like salts of two
different elements, and they certainly show none of the properties of
metallic iron, except in being slightly magnetic. But these sub-
stances must, on the present theory, be considered as giving two
different ions in solution, viz., the ferrous ion, Fe**, and the ferric
ion, Fe'". The addition of unit positive charge profoundly alters
the properties of the ferrous ion, and it is reasonable to suppose that
the sodium and chlorine atoms are also profoundly changed by the
assumption of charges by the elements. Metallic sodium, and
iron, may be regarded as discharged ions, possessing zero charge,
Na°, and Fe°. In converting an atom of iron into a ferrous ion, two
electrons are removed, producing Fe". When this is converted into
the ferric ion another electron is removed, producing Fe'". This,
however, corresponds with oxidation, since increase of positive
valency occurs. Increasing the valency of a cation therefore corre-
sponds with increasing its positive charge. Reduction is equivalent
to diminution of the positive charge on an ion, or the increase of
negative charge. Thus, ferricyanides are reduced to ferrocyanides
by increasing the negative charge on the ion by one unit : Fe(CN)6'"
-f- e = Fe(CN)6"". If iron is treated with chlorine water it forms
ferric chloride, i.e., ferric ions and chloride ions. The metallic iron
has been oxidised : Fe — 3e = Fe"a, whilst the free chlorine has
simultaneously been reduced by acquiring a negative charge :
3C1 + 3c = 3d'.

The names of the ions may be formed from the names of the
salts in which they occur, with the addition of -ion.

Fe", the ion of ferrous salts, is the ferrous-ion.
Fe'", the ion of ferric salts, is the ferric-ion.
Cr, the ion of chlorides, is the chloride-ion.
SO 4", the ion of sulphates, is the sulphate-ion.

The hydrogen-ion, H', is the ion common to all acids ; the hydroxide*
ion, OH', is the ion common to all bases.

Difficulties in the ionic theory. — The hypothesis of electrolytic dis-
sociation has still to explain how the charged atoms of, say, sodium
chloride are separated against the electrostatic forces existing
between them. Energy must be available from some source to
effect this separation, and the most reasonable assumption seems
to be that the ions are drawn apart by their attraction to molecules
of the solvent. This separation, as Larmor pointed out, must in

286 INORGANIC CHEMISTRY CHAP.

some way be effected by a steady drawing apart of the ions of each
molecule by attractive forces, the process being reversible as regards
each separate molecule, so that there is no violent disturbance,
leading to vibration and conversion of energy into heat. The actual
mechanism of ionisation, however, is still far from clear.

Another criticism advanced against the theory was that, if the
ions are free in the solution, it should be possible to separate them.
The answer to this is that such a separation can, in fact, be made. If
a layer of pure water is poured over a solution of hydrochloric acid,
the hydrogen-ions, which move more rapidly than the chloride-
ions, as we know from conduction experiments, and from direct
measurements of the speeds of ions in a potential gradient (p. 288),
will diffuse into the water. Since, however, they carry positive
charges, they will charge the water layer positively, and leave the
negative chloride-ions in the layer of acid, which thus becomes
charged negatively. By reason of the great electrostatic forces
soon set up, the hydrogen-ions tend to be dragged back into the acid,
and the chloride-ions to be pulled out, so that in a short time both
ions migrate together with equal speeds, and the acid appears to
diffuse as a whole. The existence of the electrical charges may,
however, easily be seen by placing platinum wires in the water and
in the acid, and connecting these with a galvanometer. A current
flows from the water to the acid. If a non- electrolyte, such as sugar
or alcohol, is used no trace of current can be detected.

It is not claimed that the theory of electrolytic dissociation is free
from grave difficulties. These are, however, not more numerous than
those associated with any purely chemical theory, such as that of the
structure of organic compounds, and the theory of the constitution of
benzene in particular. They are not the simple difficulties which
arise on a first acquaintaiice with the theory, such as that discussed
above, all of which are capable of ready explanation, although they are
still sometimes brought up against the theory. What can fairly be
claimed for the theory is that it has been, and still is, a valuable and
illuminating guide to research, and that it affords a consistent and simple
explanation of a large number of experimental results which otherwise
would be obscure and disconnected. All the other hypotheses proposed
in its place cover a much more restricted field, are without exception
qualitative, and in many cases incapable of experimental test. They
are, and have shown themselves to be, impotent in assisting the real
progress of scientific investigation. The great advances made in physics,
notably in connection with the elucidation of the source of the electric
current in voltaic cells, must also be kept in view. It may fairly be said
that if the theory were abandoned by chemists its position in physics
would still be assured.

The ionisation of water. — The purest water which can be obtained
is almost, but not quite, a non-conductor of electricity. After
allowing for the effects of traces of conducting impurities, a slight

xvi ELECTROLYSIS 287

conductivity, due to the ions of water itself, remains. The ionisation
of water into hydrogen-ions and hydroxide-ions is very small, and a
state of equilibrium is set up : H20 ±^: H' -f OH'. To pass a
current of 1 ampere through a centimetre cube of pure water at 18°
would require a potential gradient of about a million volts, i.e.,
the electrodes would have to be connected with 500,000 accumulator
cells in series.

The ionisation of water proceeds only to the extent of 1 gm.
mol. of water ionised in ten thousand million litres (1010 litres),
or about one-fortieth the total capacity of the earth.

Salts are electrolytes. — If 1 gm. mol. of hydrochloric acid is
dissolved in water so that the total volume of solution is 1 litre, the
conductivity of the water is increased nearly a thousand million-
fold ; 1 -2 litres of this solution contain 1 gm. of hydrogen - ions,
derived from the dissociation of the acid, whereas 1010 litres of water
contain 1 gm. of hydrogen-ions derived from the dissociation of the
water.

Most acids, bases, and salts, such as hydrochloric acid, sulphuric
acid, acetic acid, caustic potash, lime, common salt, copper sulphate,
and alum, give conducting solutions with water, and are electrolytes
(p. 278). Pure sugar, urea, alcohol, and most organic compounds,
however, do not give conducting solutions with water : they are
non-electrolytes. Since acids may be regarded as hydrogen salts, and
bases as salts containing the hydroxide radical, OH, the results
described may be summarised in the statement that salts are
electrolytes, whilst substances which are not salts are non-electrolytes.

All acids give the hydrogen-ion in solution ; all bases give the
hydroxide-ion. Dry liquefied hydrogen chloride does not redden
dry litmus, or act on zinc or marble, and it is almost a perfect insu-
lator. In solution it behaves as an acid, since then the hydrogen-
ion is formed.

The hydroxide-ion of bases, when liberated at the anode in elec-
trolysis, decomposes into water and oxygen : 2 OH = H20 -f- O.

Migration of the ions. — The bodily transfer of the ions under the
influence of an electric field can be demonstrated, and its speed
measured, by the apparatus shown in Fig. 158 (Nernst).

EXPT. 117. — The U-tube is half -filled with a solution containing 0-3
gm. of KNO3 in a litre of water. By connecting a funnel with the capil-
lary tap below the U-tube, a solution containing 0-5 gm. of KMnO4 per
litre of water, to each 100 c.c. of which 5 gm. of urea have been added to
increase its density, is slowly admitted. The surface of separation
between the colourless liquid above and the purple permanganate
solution below should be quite sharp. A current of 0-3-0-4 amp. is
now passed between the platinum electrodes, from the lighting mains.
The purple MnO4/-ions at once begin to move towards the anode, and

288

INORGANIC CHEMISTRY

CHAP.

the levels alter in the directions shown (Fig. 158). If the former levels
are marked by thin strips of gummed label, the change is quite apparent
after 10-15 minutes.

It appears from this experiment that the actual speed of move-
ment of the ions in bulk through the solution is very slow. It thus
resembles the diffusion of dissolved substances. In both cases the
moving molecules enter repeatedly into collision with the molecules
of the solvent. The actual ionic mobilities present, under a potential
gradient of 1 volt per cm., are given below in cm. per sec. (for very
dilute solutions, where the influence of ions on one another, or on
the un-ionised salt molecules, may be neglected) :

K' 0-00067 Ag' 0-00057 Cl' 0-00068 NO/ 0-00064
H' 0-00326 NaJ 0-00045 OH' 0-00181 I' 0-0069

NH4' 0-00066 S04" 0-00071

The ions in their motion are under the influence of two forces :
(i) the driving force of the potential gradient ; (ii) the viscous resistance
of the solvent. The latter frictional resistance is enormous. In
order to pull 1 gm. mol. of potassium ions
through the solution with a speed of 1 cm.
per sec. it would be necessary to apply to them
an aggregate force of no less than 1,500,000
tons (Kohlrausch).

Strengths of acids. — Since acids in solution
owe their acidic properties to the hydrogen-
ion, their relative strengths may be compared by
measuring the relative ionisations in solutions
containing equivalent weights of the acids in
identical volumes. The ionisation is most
conveniently determined by the conductivity of
the solution. Since the hydrogen-ion is much
more mobile than any of the anions of acids,
it carries most of the current, and the rela-
tive conductivities of different acids are
therefore approximately proportional to the

FIG. 158.— Demonstra- ionisations.
tion of Ionic Migration.

EXPT. 118. — One-fiftieth normal ( N/5Q) solutions

of acetic, sulphuric, and hydrochloric acids are poured into three
glass tubes, fitted with platinum electrodes, as shown in Fig. 159.
The electrodes are set at the same distance apart in the three tubes,
and in series with each tube is an ordinary carbon -filament lamp.
The tubes are connected in parallel with the lighting mains.
The lamp in circuit with the acetic acid remains dark, because
the conductivity is so small that practically no current passes. The
lamps connected with the hydrochloric and sulphuric acids light up,

xvr ELECTROLYSIS 289

but the former is brighter than the latter. The order of conductivities
of the three acids :

HC1 > H2S04 > CH3-C02H,

is therefore the same as the order of strengths found by the relative
rates of solution of zinc in the acids (p. 184).

Equivalent conductivity. — If a cell is formed containing two plati-
num electrodes 1 sq. cm. in area, placed parallel to each other at a
distance of 1 cm. apart, the current in amperes which passes through
a solution of an electrolyte between the plates, when the latter are at
a difference of potential of 1 volt, is defined as the conductivity
of the solution, and is denoted by k. Thus, conductivity =
current/ voltage for unit cube of the material.

It is found that the conductivity of a solution is very greatly depen-
dent on the concentration.

If we start with a solution containing 1 gm. equivalent of electro-
lyte per litre (e.g., HC1, or KC1, or iH2S04, or JCuS04), then we
shall have a certain number of
ions between the electrodes in the
cell, and the current carried by
these ions will be equal to the
conductivity of the solution. If
we dissolve twice as much electro-
lyte in a litre, the actual conduc-
tivity will be greater, although
there may really be a smaller
fraction of salt molecules broken
up into ions than in the more
dilute solution. Again, if we dilute
the solution containing 1 gm.
equiv. per litre to one containing

0-01 gm. equiv. per litre, the actual conductivity will be less, as there
are fewer ions between the electrodes, although a larger fraction of
salt may have been ionised. To make a fair comparison between
the ionisations of these various solutions it is evident that we must
divide the measured conductivity, k, by the number, c, of gm.
equiv. of salt per c.c. in the solution, and the quotient k/c is
called the equivalent conductivity, denoted by A. Thus A= k/c.

It is found by experiment that the equivalent conductivity of an
electrolyte increases gradually with the dilution. The curves in
Fig. 160 show the equivalent conductivities of a few electrolytes
plotted against the cube-root of the dilution in litres. It will be
seen that the curves at first rise fairly rapidly, and then slowly
approach a nearly constant value at high dilutions.

FIG. 159. — Comparison of Conductivities
of Acids.

290

INORGANIC CHEMISTRY

CHAP.

This is interpreted as follows. The ionisation of the dissolved
substance increases with dilution until, at very high dilutions, the

135
130
125
120
115
110
105
100
95
90

<85
§80

I75
| 70

!es

Jeo

.§55

ufso

45
40

0 1 2 3 4 5 6 7 8 9 1O 11 12 13 14 15 16 17 18 19 20 21 22

V~v litres
FIG. 160. — Curves showing Dependence of Equivalent Conductivity on Dilution.

electrolyte has become completely ionised. When this occurs, the
equivalent conductivity becomes constant, and this limiting
value of the equivalent conductivity, corresponding with complete

XVI

ELECTROLYSIS

291

ionisation, is denoted by A^ (i.e., the value at infinite dilution).
Since there are now only ions in the solution, the ratio A'/c, or A,
has become constant.

Thus, if -we have 1 gm equiv in 106 litres practically completely
ionised, giving a certain conductivity klf and we then dilute the solution
to 1010 litres, we obtain a smaller conductivity, &2. But if we suppose
all the ions present to be collected into 1 c.c. in each case, it is evident
that we should have two identical solutions, since the numbers of ions
are equal, and thus A is the same for both This would not be true
except when the salt is completely ionised, i e , at very great dilutions,
because then the number of ions in the two solutions considered would
be different, and if we brought them all into 1 c c. the conductivity
in one case would be different from that in the other.

Degree of ionisation. — The ratio of the equivalent conductivity
at any dilution, v, to that at infinite dilution, i.e., to the limiting
conductivity for infinite dilution when all the electrolyte is ionised,
is the degree of ionisation, a, corresponding with the given dilution
of the solution : A^/A^ = a. By the dilution is understood the
reciprocal of concentration, i.e., the number of c.c. containing 1 gm.
equivalent of total electrolyte.

In practice, the concentration is usually measured in gm. equiv.
per litre, and the dilution in litres per gm. equiv. In these units
A = (k/c') X 1000, or (kv') X 1000.

The progressive ionisations of two typical electrolytes are seen
from the tables below.

IONISATION OF KC1 AT 18

c gm. equiv.
per litre.

0-0001

0-001

0-01

0-1

1-0

Equivalent

conductivity

A=(fc/c)xlOOO

129-9
129-1
127-3
122-4
112-0
98-3

Degree of
ionisation

1-00

0-994

0-980

0-943

0-862

0-757

Ionisation
constant K

0-0154
0-0485
0-1542
0-5405
2-350

IONISATION OP ACETIC ACID AT 18°

Dilution v
litres per gm.

Equivalent
conductivity

Degree of
ionisation

equiv.

A=Jcv x 1000

o = A/AOO

0-334

0-6186

0-0016

1-977

2-211

0-0057

10-753

5-361

0-0138

63-26

13-03

0-0336

387-9

1-0000

7-7
16-5
18-0
18-5

Ionisation
constant
K
X 10-6

x io-6

X 10-«
X 10~6

"u"2

292

INORGANIC CHEMISTRY

CHAP.

Potassium chloride is appreciably ionised even in normal solution :
it is a typical strong electrolyte. Acetic acid is only slightly ionised,
even in dilute solutions (when completely ionised it has a higher
equivalent conductivity than potassium chloride, owing to the great
mobility of the hydrogen-ion) ; it is a typical weak electrolyte. The
significance of K will be considered later (p. 357).

Determination of conductivity. — If an ordinary current from a
battery is passed between platinum electrodes in a solution of an
electrolyte, and a galvanometer is included in the circuit, it will be
found that the strength of the current diminishes as electrolysis
proceeds. This diminution in current strength is partly due to "the
accumulation of the products of electrolysis at the electrodes. These
form a galvanic cell which tends to send a current in the opposite
direction to that driven round the circuit by the battery.

This reverse electromotive force, tending to oppose the direct
electromotive force of the battery which is effecting decomposition,
is known as the electromotive force of polarisation.

In order to obtain accurate measurements of the conductivity
of electrolytes it is necessary to eliminate polarisation. F. W,
Kohlrausch (1869) did this by using an alternating current, i.e.,
a current which flows alternately in one direction and then in the
other, with a very small interval of time between the reversals
of direction. Such a current is supplied by an induction coil
(without condenser) attached to a battery.

The ions are driven first in one direction and then in the other by
the alternating current, and the amounts
deposited on the electrodes are exceedingly
small.

Polarisation is still further reduced by deposit-
ing platinum black on the electrodes, by electro-
lysing between them a solution of 1 gm. of
chloroplatinic acid and 8 mgm. of lead acetate in
30 C.c. of water, with an accumulator, and
reversing the current from time to time.

EXPT. 119. — A convenient type of electrolytic
cell is shown in Fig. 161. It consists of a small
stoppered bottle (shown full size), with parallel
platinised platinum electrodes sealed in. The
platinum wires from the electrodes, which a,re
covered with glass inside the cell, pass into

glass tubes on each side. A drop of mercury is poured into each
tube, and the wires from the coil dip into the mercury to make contact.
These wires then pass through rubber tubes, so that the cell, filled with
a solution of KC1, say decinormal, and stoppered, may be immersed
in a tank of water kept at a constant temperature, say 18° or 25°.

FIG. 161.— Cell for Measure-
ment of Conductivity.

XVI

ELECTROLYSIS

293

The alternating current is supplied by a small induction coil giving
a high buzzing ncte, with one accumulator. Since a galvanometer
cannot be used with an alternating current, a telephone is employed.

FIG. 162. — Apparatus for Measurement of Conductivity.

In order to measure the resistance, a resistance -box is connected
with the conductivity cell, telephone, and coil, and a metre wire-
bridge with a scale and sliding
contact. Fig. 162 shows the
apparatus set up for use. The
connections are shown in Fig.
163.

The slider is placed near the
middle of the bridge and plugs
are taken out of the resistance
box until the sound in the tele-
phone is appreciably reduced.
The slider is then moved about
until the sound in the tele- FIG. 163.— Diagram of Conductivity Apparatus,
phone is reduced to a minimum.

Let a be the reading on the bridge, R the resistance taken out of the
box : then the resistance of the conductivity cell, r, is, since the arrange-
ment constitutes a Wheatstone bridge, given by r = R X -r^ ohms.

luu — a

The conductance is 1/r, i.e., the current passing, in amperes, for 1 volt
potential difference between the electrodes. This follows from Ohm's

law :

potential difference in volts

Current m amperes = r- — - -T-- - -•»

resistance in ohms

which has been proved experimentally to apply to electrolytes.

The electrodes of the conductivity cell will not usually be exactly
1 sq. cm. in area, parallel, and 1 cm. apart, so that the conductance is not
usually equal to the conductivity (p. 289). Since the relation between
th«» two depends only on the construction of the cell, it is possible to
<l»'t»M-Miino once for all this ratio, called the cell constant, for a particular
coll. This is done by losing as electrolyte a solution of known conduc-

294

INORGANIC CHEMISTRY

CHAP.

tivity : a normal solution of potassium chloride (74-55 gm. per litre),
for which at 18° &18Q =•= 0-09824. If the resistance of the cell containing

x-y

this solution is r ohms, &18o = — = 0-09824, where C is the cell constant.

If any other solution is now used, and if the resistance is r' ohms, the
conductivity is fc'18° = C/r'.

Neutralisation. — Acids are substances producing the hydrogen-ion
in solution : HC1 ^± H* -f- Cl'. Bases are substances producing
the hydroxide-ion in solution : NaOH ^± Na" -f OH'.

If an acid and a base in solution are mixed, a salt is formed, and
the solution becomes neutral. This is usually represented by such

Accumulator

FIG. 164. — Apparatus to demonstrate Diminution in Conductivity on Neutralisation.

equations as : HC1 -f- NaOH = NaCl -f H2O. Since the acid, base,
and salt are usually ionised in solution, the reaction really occurs
between the ions : (IT + Cl') + (Na' + OH') = (Na' + Cl') +
H20. It will be seen that the anion of the acid (Cl'), and the cation
of the base (Na"), which together constitute the ions of the salt,
take no part in the change : they are free before and after the reaction.
The net change in neutralisation is the union of the hydrogen-ion of
the acid with the hydroxide-ion of the base to form practically undisso-
ciated water : Hr -f OH' = H20.

This is the sole reaction with strong acids and bases, i.e., those
which are practically completely ionised. Salts are nearly always
largely ionised in solution.

The hydrogen- and hydroxyl-ions are those which possess the
greatest mobility (p. 288). After neutralisation, therefore, when the
rapid hydrogen- and hydroxyl-ions have been withdrawn, the con-
ductivity of the solution will be appreciably diminished.

EXPT. 120. — Fit a rectangular glass trough with two electrodes of
sheet copper (Fig. 164). Connect these through an ammeter with
two accumulators in series. Pour into the cell a solution of A7-caustic
soda containing dissolved urea to increase its density, and coloured with

xvi ELECTROLYSIS 295

phenolphthalein. Float a slice of cork on this solution and, by means of
a pipette, introduce an equal volume of AT-hydrochloric acid as a definite
stratum above the alkali. Switch on the current and observe the
deflection of the ammeter. This is a measure of the current carried
by all the ions, Na', H', OH', Cl'. Now stir the two liquids with a glass
rod, and notice the reduced reading of the ammeter. The ions Na'
and Cl' alone now carry the current.

A modification of this method may be used in titrating an alkaline
or acid solution which is too strongly coloured to allow of an indicator
being used.

Heat of neutralisation. — If the theory of neutralisation given above
is true, the heat evolved in the neutralisation of one equivalent of
a strong base by one equivalent of a strong acid should be the same
for different acids and bases, since the reaction in all cases is the
same, viz., the union of hydrogen-ions from the acid with hydroxide-
ions from the base to form practically undissociated water.

This unexpected result has been verified by experiment ; the heat
of neutralisation is, per equivalent of strong acid and base, equal to
about 13-7 kgm. cal.

HC1 Aq + NaOH Aq 13-70 HNO3 Aq + NaOH Aq 13-70
HBr Aq + KOH Aq 13-76 HC1 Aq + |Ba(OH)2 Aq 13-80.

If the acid or the base is weak, heat will be evolved or absorbed
during the neutralisation, since the un-ionised acid or base will
dissociate as neutralisation proceeds, and hydrogen-ions and hydr-
oxide-ions are removed ; this dissociation will, in general, be attended
by an absorption or evolution of heat. An example of this behaviour
is the neutralisation of hydrofluoric acid (p. 421). If the salt formed
is only slightly ionised (a case which is very rare), or is insoluble,
the heat of neutralisation will also be abnormal, since the association
of the ions of the salt to form undissociated molecules, or the precip-
itation of the salt, are processes attended by heat changes.

lonisation in stages.— Molecules which are capable of giving more
than two ions on dissociation often dissociate in stages. This is
not always the case. Thus, potassium ferrocyanide, K4Fe(CN)6,
ionises directly accoroling to the equation : K4Fe(CN)6
= 4K' -f Fe(CN)6'"', whilst sulphuric acid ionises in two stages :
H2S04 '= IT -f HS04', followed by HSO4' = H' + SO4". The
second stage of the dissociation occurs only to a very limited extent,
except in very dilute solutions.

At moderate dilutions, therefore, sulphuric acid should behave as
a monobasic acid. The conductivity shows that this is the case.
But if the acid is neutralised with caustic soda, the hydrogen-ion
is completely eliminated, with the hydroxide-ion of the base, in the
form of water : H2S04 + 2Na' -f 20H' = 2Na* + SO4" + 2H2O.

296

INORGANIC CHEMISTRY

CHAP.

The reason for this behaviour is the ionisation of the HS04'
ion into H' and S04". As soon as the hydrogen-ion corresponding
with the first stage of the ionisation : H2S04 = H' -f HSO4',
has been removed, the HSO4' ion begins to dissociate to a slight
extent. The trace of hydrogen-ion so produced, however, is at
once removed by the hydroxide-ion of the base added, and further
ionisation of HS04' results. This goes on until all the HS04' has
been ionised, and finally only SO/7 ions remain. This, however,
corresponds with the formation of the normal salt, and the acid,
therefore, behaves as if it were dibasic.

Manufacture of chlorine and alkali by electrolysis. — The electro-
lysis of brine, i.e., a solution of sodium chloride, studied in
Expt. 116, is applied on a large scale for the manufacture of caustic
soda and chlorine.

The Castner-Kellner cell consists (Fig. 165) of a shallow slate tank
divided into three compartments by slate partitions not quite

touching the floor. The
C floor is covered by a pool

of mercury, thus separ-
ating the three com-
partments. Each end
compartment is filled
with strong brine, the
middle one with water.
Anodes of carbon, as
shown, or of platinum
gauze, are placed in the
end compartments, whilst
the cathode consists of a
bundle of iron rods in the
central compartment. Chlorine is evolved in the end compartments,
and is led off by earthenware pipes. Sodium ions are discharged
on the mercury in the end compartments, and the sodium dissolves
in the mercury, forming sodium amalgam. The cell is given a slow
rocking motion by an eccentric, and the sodium amalgam is brought
from the end compartments to the middle compartment, where it
decomposes the water, forming a solution of caustic soda. Hydro-
gen is evolved from the iron cathode. In the new type of cell the
tank is stationary, and the mercury is moved by an archimedean
screw, finally dropping over a cascade into water to free it from
sodium, after which it re-enters the cell. The Castner-Kellner cell
is used at Niagara ; a smaller plant is in operation at Weston Point,
near Liverpool.

The Electro-Bleach Company, at Middlewich, in Cheshire, use
the Hargreaves-Bird cell. The cell consists of a narrow rectangular
box (Fig. 166), the top and bottom of which are of cast iron, and the

FIG. 165.— Castner-Kellner Cell.

ELECTROLYSIS

297

sides of asbestos-board soaked in sodium silicate to act as diaphragms.
Outside the diaphragms, and in contact with them, are the cathodes
of copper gauze. The anodes are inside the chamber, and consist of
five lead cores, on which are strung rough
blocks of gas-carbon, the exposed lead being
covered with cement. Brine is fed in at the
bottom of the cell, and the spent liquor, still
containing about one-third of the salt undeconv
posed, runs off to waste from the top. The
sodium ions pass through the diaphragms, and
are discharged on the cathodes. Steam is
blown on to these, which are enclosed in an
outer iron casing, and a solution of caustic
soda obtained. Chlorine escapes from a pipe
at the top of the anode chamber. The gas
may be diluted with air, and used in the
manufacture of bleaching powder, or liquefied
by compression to 6 atm. at 15°, or by cooling
at ordinary pressure in iron pipes. It is sent
out as liquid chlorine in steel cylinders. The
electrolytic chlorine is purer and more concen-
trated than that made by chemical methods. Some of it is used
in making stannic chloride, or chlorinated acetylenes (p. 680).

==]) 1

i ^

1
1

FIG 166._Hargreaves-
Bird Ceil.

SUMMARY OP CHAPTER XVI

Faraday's Laws of Electrolysis: (1) The weight of an ion deposited
in a given time is proportional to the strength of the current ; (2) 96,000
coulombs _ liberate 1 gm. equiv. of any ion. This quantity of elec-
tricity, 96,000 cmb., is called a faradayj denoted by F.

Theory of Electrolytic Dissociation : Salts in solution are dissociated
into electrically-charged atoms or radicals, called ions. These ions carry
the current in electrolysis. The charge on an ion is either positive
(cation), or negative (anion), and is always equal to the fundamental
charge, c , multiplied by the valency of the ion. The unit charge, f, is
identical with the charge of the atom of free negative electricity, or
the electron. Its value is 1-59 X 10~19 cmb. Free positive electricity
is not known ; a positive ion is an atom or radical which has lost one
or more negative electrons. The conductivity of an electrolyte is the
current in amperes which passes through the solution contained in a
cubical cell with sides 1 cm. long when the opposite sides, forming
electrodes, are connected with a potential difference of 1 volt. The
equivalent conductivity of a solution is the conductivity divided by the
concentration in gm. equiv. per c.c. : A = k/c. It increases with
dilution, since the ionisation increases. The degree of ionisation
measured by the ratio of the equiv. conduct, at a given concentration
to the equiv. conduct, at infinite dilution, corresponding with complete
ionisation, «• = A.A .

298 INORGANIC CHEMISTRY CH. xvi

Neutralisation of a strong acid by a strong base is the union of the
hyoirogen-ion of the acid with the hydroxide-ion of the base to form
undissociated water. The ionisation of water is very small.

EXERCISES ON CHAPTER XVI

1. Describe what happens when an electric current is passed between
platinum plates through solutions of (a) copper sulphate, (6) sodium
sulphate, (c) potassium chloride. Distinguish between primary and
secondary products of electrolysis.

2. How is caustic soda manufactured by electrolysis from common
salt? In order to decompose 1 kgm. of salt, how many ampere-hours
(1 ampere flowing for 1 hour) are theoretically necessary?

3. State Faraday's Laws of Electrolysis. What experiments would
you make in order to demonstrate the truth of these laws?

4. Give a short account of the theory of electrolytic dissociation,
and indicate what explanations it gives of (a) electrolysis, (b) the heat
of neutralisation of a strong acid by a strong base.

5. Represent by ionic equations the following reactions : (a) the
solution of magnesium in hydrochloric acid, (b) the action of water on
sodium, (c) the preparation of chlorine from manganese dioxide and
hydrochloric acid, (d) the precipitation of silver nitrate solution by a
solution of sodium chloride.

6. What are electrons ? How has the absolute value of the charge of
an electron been determined ? From this, how is it possible to calculate
the number of molecules of hydrogen in 1 c.c. of hydrogen gas at S.T.P. ?

7. What weight of silver is deposited from a solution of silver nitrate
by a current of 0-075 ampere flowing for seventeen minutes ?

8. Describe how the speed of migration of an ion may be measured.
How does it compare with the speed of diffusion of a gas, and what
connection is there between the two?

9. What is the degree of ionisation of an electrolyte ? How may it
be measured, and how does it change with dilution ?

10. A current of 0-1 amp. is passed for forty -five minutes through a
voltameter containing acidulated water, and one containing copper
sulphate solution with copper electrodes. What volume of electrolytic
gas (at S.T.P.) will be evolved, and what weight of copper deposited ?

CHAPTER XVII

THE MOLECULAR WEIGHTS OF SUBSTANCES IN SOLUTION

The molecular depression of freezing point. — The lowering of the
freezing point of a solvent by a substance in solution is proportional
to the concentration of the latter (p. 103). Thus, with cane-sugar
in water :

Gm. of sugar Freezing point

in 100 grams lowering Ratio

of water = C. = D. D/C.

11-4 0-62° 0-0544

22-8 1-23° 0-0544

34-2 1-85° 0-0544

The depression of freezing point produced by 1 gm. of sugar in
100 gm. of water is 0-0544°; that by n gm. is 0-054471°.
The depression produced by 1 gm. of another substance will be
different, e.g., 1 gm. of urea lowers the freezing point of 100 gm.
of water by 0-31°.

Raoult (1883) made the important discovery that if quantities
proportional to the molecular weights of different substances are dis-
solved in identical weights of a solvent, the freezing points of all the
solutions are the same. A molecular weight in grams of a substance
dissolved in 1 kilogram of water depresses the freezing point of the
latter by 1-85°. This is called the molecular depression of freezing
point, A , for water.

Thus, if 342 gm. of cane-sugar, corresponding with the formula
C^HagOj!, are dissolved in 1 liore of water, the solution freezes at
— 1-852°. The same freezing point is shown by solutions of 60 gm. of
urea, CON2H4, or 46 gm. of alcohol, C2H6O, in 1 litre of water, since
these are equi-molecular amounts.

The molecular depression varies with the solvent. Thus, if
46 gm. of absolute alcohol are dissolved in 1 kilogram of benzene,

299

300 INORGANIC CHEMISTRY CHAP.

CgFfi, the solution freezes at 0-5°. The freezing point of pure benz-
ene is 54°, hence the molecular depression of freezing point for
that solvent is 4-9°. The values of the molecular depressions for
some common solvents are as follows :

A M. pt. A M. pt.

Water 1-862° 0° Formic acid 2-8° 8°

Acetic acid 3-9° 17° Phenol 7-27° 40°
Benzene 4' 9° 5°

Van't Hoff (1886) showed that A may be calculated from the latent
heat of fusion, Lf, and the absolute melting point, Tf, of the solvent,

0 -009 7*2

by the formula : A = " • For water : Lf = 79-77, Tf = 273,

:. A = 0-002 X (273)2/79-8 = 1-869.

It is clear that a measurement of the freezing point of a solution
enables us to find the molecular weight of the dissolved substance, in
the state in which it exists in solution.

Let the depression of freezing point produced by m gm. of solute
per kgm. of solvent be D. That produced by the molecular weight,
M, in 1 kgm. we know is the molecular depression A. Further, we
know from Blagden's law that the two depressions are proportional
to the two concentrations :

.'. m : M : : D : A

hence M = m X j? '

EXAMPLE. — 1-35 gm. of carbon tetrachloride were dissolved in 55
gm. of acetic acid. The freezing point of the latter was depressed from
16-750° to 16 '132°. Find the molecular weight of carbon tetrachloride.

m = No. of gm. of solute per 1000 gm. of solvent = 1-35 X 1000/55

Observed depression = 16-750 — 16-132 = 0-618° = D.

Molecular depression for acetic acid = 3-9° = A.

Molecular weight of solute M =2^ = 1^5x1^0x^9 = ^

D 55 X 0-bl8

The molecular weight calculated from the vapour density is CC14 =
153, hence we conclude that carbon tetrachloride has the same molecular
weight in the state of vapour as in solution in acetic acid ; in both cases
the formula is CC14.

Raoult's law holds good only if the solution is dilute ; apparent
exceptions are also shown by aqueous solutions of acids, bases,
and salts (i.e., electrolytes) ; these correspond with the ionisation
of the substances. In its application to the determination 01 mole-
cular weights, two conditions must therefore be satisfied : (i) the
solution must he dilute, and (ii) the solution must not be an electrolyte.

xvn MOLECULAR WEIGHTS OF SUBSTANCES IN SOLUTION 301

Determination of molecular weights by the freezing-point method.—

The apparatus used in the determination of molecular weights from

the depression of freezing point is shown in Fig. 167. A very

sensitive thermometer, called a Beckmann

thermometer, D, is used, which has only

six degrees on the whole scale, the latter

being graduated in thousandths of a

degree. There is a reservoir at the top of

the capillary tube, into which mercury

can be shaken if higher temperatures are

used (e.g., phenol, m. pt. 40°), or from

which mercury can be drawn into the

tube and bulb if lower temperatures (e.g.,

water, 0°) are to be used. It is of no

consequence what the actual readings on

the scale are, it is only their difference,

D, which is required. The solvent is

weighed into the tube A, and a stirrer of

bent platinum wire introduced. The

thermometer is now fitted into the tube

through a cork, so that the bulb is

covered with the liquid.

The tube A is then fitted through a cork
into a large test-tube, B, which serves as
an air-jacket, and prevents the fall in
temperature being too rapid. The tube B
is plunged into a freezing mixture (e.g., ice
and salt) contained in the large jar, C.
The stirrers in the solvent tube and outer
jar are worked up and down, and the
thermometer observed. The mercury falls
steadily to a certain point, when the
solvent is slightly supercooled. Freezing
-then commences, the temperature at
once runs up to the freezing point, and
afterwards remains stationary. It is then
read off with a lens, the thermometer being
gently tapped to prevent any adhesion of
the mercury to the glass. Suppose the
reading is 3-216°.

The tube A is then taken out, and
allowed to warm until the solvent

liquefies. A weighed quantity of the substance under investiga-
tion is introduced through the side tube, and dissolved by
working the stirrer. The tube is replaced in the air-jacket, and the
latter again put into the freezing mixture. The process is carried

FIG. 167. — Beckmann's
Freezing-point Apparatus.

302 INORGANIC CHEMISTRY CHAP.

out further exactly as with the pure solvent, and the freezing point
read off. Suppose this to be 2-839°; then D, the depression of
freezing point, is 3-216 - 2-839 = 0-377°.

A mixture of ice and salt is used in the outer jar if the solvent
is water ; ice and water are used for benzene, acetic acid, and formic
acid ; phenol is melted in warm water, and the inner tube and air-
j acket are supported in a clamp without outer j ar . Acetic and formic
acids, and phenol, readily absorb moisture, which lowers their
freezing points. Care must be taken to prevent this occurring
during the experiment.

EXAMPLE. — 17-79 gm. of an aqueous solution containing 0-1834 gm,
of hydrogen peroxide froze at — 0-571°, What is the molecular weight
of hydrogen peroxide in the solution ?

0-1834 gm. of hydrogen peroxide is dissolved in 17-79 — 0-183 =
17-607gm. of water

, 0-1834 X 1000
.'. wt. of peroxide in 1 kgm. of water = TT^ROT — == *"'** Sm- ~ m>

Let M = mol. wt. of hydrogen peroxide, then :
10-42 :M = 0-571 : 1-86

:.M = 10-42x^-33-9.
The formula H2O2 gives M = 34.

Vapour pressures of solutions.— It has already been mentioned
(p. 104) that salt, when dissolved in water, lowers the vapour pressure
of the latter. This is quite general : if a non- volatile substance is
dissolved in a volatile solvent, the vapour pressure of the solution is
lower, at a given temperature, than that of the pure solvent. Further,
if /0 is the vapour pressure of the pure solvent, / that of the solution,
/ f /\

the ratio ( ^7— - ), or the relative lowering of vapour pressure, is

\ Jo /
found to be (1) proportional to the concentration of the solution ;

(2) practically independent of temperature within certain limits ;

(3) the same for equimolecular amounts of different substances in
the same weight of a solvent. (Raoult, 1887.)

The molecular lowering of vapour pressure is therefore a constant
for a given solvent. In a solution containing N0 gm. mol. of solvent
and N gm. mol. of solute, the relative lowering of vapour pressure is
found by experiment to be given by the equation :

Thus, if 1 gm. mol. of solute is dissolved in 99 gm. mol. of solvent,
there will be a lowering of vapour pressure of 1 per cent., since
N/(N0+ N) = l/(-99 + 1) = 0-01. The value of N0 is calculated
from the weight of solvent taken divided by its molecular weight in
the state of vapour, i.e., NQ is the number of vapour molecules.

pmm

760

xvn MOLECULAR WEIGHTS OF SUBSTANCES IN SOLUTION 303

EXAMPLE. — Pure benzene, C6H6, has a vapour pressure of 751-86
mm. at 80°. When 2 '47 gm. of ethyl benzoate are dissolved in 100
gm. of benzene, the solution has a vapour pressure of 742-6 mm. The
molecular weight of benzene vapour is 78 /. N0 = 100/78 = 1-283.
Also (/0 -/)//o = (751-86 - 742-6)/751 -86 = 0-0123 /. 0-0123 =
N/( 1-283 + N) .'. N = 0-01598. But N = 2-47/(mol. wt. of ethyl
benzoate) /. mol. wt. of dissolved ethyl benzoate = 2-47/0-01598
= 154*6. That calculated from the vapour density, or the formula
C6H6-COO'C2H5, is 150.

Boiling points of solutions. — Lowering of vapour pressure is
synonymous with elevation of boiling point, since the latter is the
temperature at which the vapour pressure reaches atmospheric
pressure, or 760 mm. The boiling point of water is 100°, because
at 100° the vapour pressure of pure water is 760 mm. If salt is
dissolved in the water, the vapour pressure at 100° is less than 760

mm., and it will be necessary to
raise the temperature above 100°
to attain that pressure, i.e., the
boiling point of the water is raised
by the dissolved substance. Let
A A (Fig. 168) be the vapour
pressure curve of the pure
solvent, BB that of the solution.
Since the relative lowering of
vapour pressure for a given
concentration is independent of
temperature between certain
limits, the curve BB will be
at a constant distance from
the curve A A. If we draw

a horizontal line through p = 760 mm. it cuts the curves at points
corresponding with the boiling point of the pure solvent, and that of the
solution, respectively, viz., t0, tv If we have a still more concentrated
solution, the vapour pressure curve of which is CC, the boiling point
is tz, and t2>t1>tQ. The vapour pressures at the temperature t0
are tQa, t0b, t0c, respectively. If cv c2 are the concentrations of the
two solutions, the lowerings ab, ac are proportional to cl5 c2 (since /0
is constant). But, since the curves are parallel, ah : ac : : ae : ah,
i.e., the elevation of boiling point is proportional to the lowering of
vapour pressure, and both are proportional to the concentration :

% : 8p2 : : 8^ : &a : : Cj : C2.

The elevation of boiling point of a solution is often applied in the
laboratory to produce a heating-bath of higher temperature than 100°.
For this purpose, solutions of the very soluble salt calcium chloride are

FlO. 168. — Vapour Pressure Curves of
Solutions.

304 INORGANIC CHEMISTRY CHAP.

convenient. They may be boiled in iron vessels. The boiling points
for given amounts of anhydrous salt are as follows :

Parts of calcium chloride per 100 parts of water : 50 200 325
Boiling point : ... 112° 158° 180°

Such high-temperature baths may replace those using oil, glycerin,
or fusible metal, except at temperatures above 200°.

It follows that the molecular elevation of boiling point is constant
for a given solvent. It is taken as the rise in boiling point for 1
gm. mol. of non- volatile solute in 1 kgm. of solvent ; we may denote
it by E.

If w gm. of substance in 1000 gm. of solvent raise the boiling
point D°, we shall have the proportion D : E : : w : M, where M is

the molecular weight of the dissolved substance. Hence M — -g-

(c/. the freezing-point equation : M = &w/D, p. 300).
The values of E for a few solvents are given below.

Boiling Molecular elevation of

Solvent. point °C. Boiling point, E-.

Water 100 0-52

Chloroform 61-2 3-66

Methyl alcohol 64-7 0-88

Ethyl alcohol 78-3 1-16

Ether 35-4 2-10

Benzene 80-2 2-57

The value of E may be calculated from the latent heat of evaporation
of the solvent, Le, in a similar way to that of A from the latent
heat of fusion. If T is the absolute boiling point,

,, _ 0-002T2

E, — = •.

•Lie

Thus, for water : T = 100 -f 273 = 373 ; Le = 538

.'. E = 0-002 X (373)2 -h 538 = 0-517 (obs. 0-52).
The above equation does not hold for concentrated solutions, or for
solutions of electrolytes (p. 300). It applies to many organic substances
(e.g., sugar, urea) in water, and in organic solvents.

EXAMPLE. — The molecular weight of iodine dissolved in ether may be
calculated from the following figures :

2-0579 gm. of iodine dissolved in 30-14 gm. of ether gave an elevation
of boiling point of 0-566°.

w = 2-0579 X 1000/30-14 = 68-28 ; D = 0-566° ; E = 2-10°

.'. M = Ew/D = 2-10 x 68-28/0-566 = 253-3. But I2 = 2 x 126
= 252, .'. iodine exists as diatomic molecules, I2, in solution in ether.

xvn MOLECULAR WEIGHTS OF SUBSTANCES IN SOLUTION 305

Determination of the elevation of boiling point. — The apparatus
for the determination of the molecular weight of a dissolved sub-
stance from the elevation of the boiling point of a solvent, devised
by Beckmann, is shown in Fig. 169. The thermometer and the
tube for holding the solution are the same as those used in the
freezing-point apparatus (p. 301), except that the tube has a short
piece of platinum wire sealed through the bottom to assist in the
transmission of heat, and contains a layer of small crystals of garnet
to prevent bumping. The
tube is surrounded by a
glass mantle, plugged
with asbestos, and the
closed end of the tube,
with the platinum wire
projecting, is heated over
a slightly smaller hole in
a piece of asbestos mill-
board, with wire-gauze
beneath, by means of a
small Bunsen flame, so
as to get a uniform tem-
perature. The vapour of
the solvent formed in
the tube is condensed in
a reflux condenser, most
conveniently formed of a
limb of the tube, as
shown, and the liquid
flows back. The boiling
point of the pure solvent
is first found in terms
of an arbitrary reading
on the Beckmann thermo-
meter, the mercury
column in the latter
having been suitably ad-
justed. A weighed quan-
tity of solvent is used

for this purpose. The tube is then cooled, and a weighed quantity
of the substance, the molecular weight of which is to be found, is
introduced through the side tube and completely dissolved. The
boiling point of the solution is then found. The difference is D,
the elevation of boiling point.

A more convenient apparatus is that of Landsberger, modified
by Beckmann (1902), in which the solution is heated by passing
through it the vapour of the solvent. The latter condenses, giving

FIG. 169. — Beckmann's Apparatus for Determination
of Elevation of Boiling Point.

306 INORGANIC CHEMISTRY CHAP,

out heat and raises the temperature of the solution (which is, of
course, becoming diluted), until the boiling point is reached. The
vapour of the solvent then passes through without condensation.

The glass tube A
(Fig. 170) contains the
solvent, and is heated
in the same way as in
the preceding appar-
atus. Inside is the
tube B, graduated in
mm., containing the
solution and the
thermometer. Vapour
from A bubbles
through the solution
by way of the tube O,
open to the vapour mA. The tube R
prevents liquid from B being sucked back
into A. The vapour is condensed in E,
and the liquid can be allowed to flow
back into B, or, by turning the condenser
in the ground joint, returned to A
through the siphon-tube F, the opening
of which is brought opposite a hole
shown. The concentration of the solution
when vapour passes freely through it and
the boiling point remains constant is
determined by weighing.
Deliquescence. — If a beaker containing pure water and one con-
taining a solution of a salt, or other non- volatile substance, in water
are placed side by side under an evacuated receiver (Fig. 171),
each liquid emits aqueous vapour into the space
above. If the vapour pressures were the same,
equilibrium would be established with a definite
pressure of vapour in the space — this, in fact,
occurs when either liquid is separately confined
under the receiver. In this state as many
molecules of water are leaving the liquid per
second by evaporation as are returning to it by
condensation. But the vapour pressure of the
pure water is always higher than that of the
solution, hence the pure water tends to
saturate the space with vapour under a
higher pressure than can remain in equilibrium with the solution.
Condensation of vapour occurs on the latter, and the pure
water is gradually transferred completely to the solution by this

FIG. 170. — Landsberger Boiling-
point Apparatus — Beckmann .

FIG. 171.— Isothermal
Distillation.

xvi i MOLECULAR WEIGHTS OF SUBSTANCES IN SOLUTION 307

process of isothermal distillation, the solution becoming diluted.
Equilibrium is set up when all the pure water is evaporated and
absorbed by the solution, and a little aqueous vapour exists in the
space, under a pressure equal to the vapour pressure of the diluted
solution.

Many solid salts, such as potassium carbonate and calcium chlor-
ide, become damp on exposure to moist air, and in time liquefy
completely. This liquefaction of solids on exposure to moist air is
called deliquescence. All deliquescent substances are very soluble
in water, and this suggests the explanation of the change. If a
trace of moisture is present in the solid, a little saturated solution is
formed. Since this is very concentrated, its vapour pressure can be
less than the partial pressure of aqueous vapour in the atmosphere,
however low the latter may be. Moisture is attracted by the salt,
which gradually liquefies completely to a saturated solution. The
latter then goes on absorbing aqueous vapour until its dilution is
such that the vapour pressure is equal to the partial pressure of
water vapour in the air.

Solid substances which attract moisture without liquefaction,
such as recently-ignited charcoal, and li quids such as alcohol and
sulphuric acid which absorb moisture, are called hygroscopic. In
the first case the moisture appears to be condensed on the surface of
the charcoal by molecular attraction (cf. p. 270). A similar cause
may explain the commencement of the deliquescence of recently-
fused calcium chloride, caustic potash, etc.

Osmotic pressure. — If a concentrated solution of copper sulphate,
contained in the lower part of a cylinder, is covered with a layer of
water, the copper sulphate molecules gradually diffuse upwards
until the solution becomes homogeneous, and of uniform colour
(p. 258). The dissolved molecules thus behave to some extent like
those of a gas ; in both cases they are in motion, i.e., they possess
kinetic energy. If we could interpose a partition in the solution, with
pure water above, which would stop the dissolved copper sulphate
molecules from passing, but would be freely permeable to water,
we should expect the copper sulphate molecules to exert a bom-
bardment pressure on the partition. A partition which is freely per-
meable to pure solvent, but is impermeable to dissolved substances,
is called a semipermeable partition, or — since it is usually prepared
in the form of a thin film — a semipermeable membrane.

A semipermeable membrane may be regarded as a kind of
molecular sieve, or filter. Just as an ordinary filter-paper will stop
suspended particles, and permit dissolved molecules to pass through,
the semipermeable membrane may be regarded as stopping even the
dissolved molecules, and permitting only the molecules of pure
solvent to pass through. There is, however, a difference between
the two cases : in order to squeeze pure solvent through the semi-

x 2

308

INORGANIC CHEMISTRY

CHAP.

permeable membrane it is necessary to apply a definite, and often
large, pressure to the solution enclosed in it. At lower pressures no
solvent percolates through the partition.

Various substances have been discovered which function as semi-
permeable membranes. Without exception they are slimy, non-
crystalline bodies, called colloids (p. 314). Thus, if a drop of copper
sulphate solution is introduced into a solution of potassium ferro-
cyanide from a pipette, a skin or pellicle forms over it, composed of

copper ferrocyanide, Cu2FeC6N6. This sub-
stance is produced as a reddish-brown,
gelatinous precipitate when the two solu-
tions are mixed : 2CuS04 + K4FeC6N6 =
Cu2FeC6N6 + 2K2SO4. The pellicle is
semipermeable, because if we allow the

Jdrop to stand in the solution, no copper
salt diffuses through, as may be seen
from the ferrocyanide solution remaining
clear. The drop usually, however, ex-
pands or shrinks, owing to passage of
water in or out through the pellicle. By
holding the drop suspended, with a bright
light behind the beaker, the streaks
due to changes of concentration may be
seen.

EXPT. 121. — Into a strong solution of
sodium silicate place small pieces of ferric
chloride, nickel chloride, cobalt chloride, and
copper chloride. Observe the formation of
pellicles, which assume curious shapes on
standing. ("Chemical Garden.")

Measurement of osmotic pressure. — In

order to give strength to the membrane,
so as to make it capable of withstanding
considerable pressures, Pfeffer in 1877
deposited the copper ferrocyanide in the
walls of an unglazed earthenware cell,
such as is used for the porous pots in
galvanic batteries.
The pot is immersed in copper sulphate solution, and placed
under the receiver of an air-pump. The air in the pores is then
removed, and on admitting air to the receiver, the copper solution is
forced into the pores of the pot. The latter is removed from the
solution, quickly washed out, and filled up with a 3 per cent, solution
of potassium ferrocyanide. The pot is then allowed to stand for

FIG. 172. — Pfeffer's Apparatus
for Measurement of Osmotic
Pressures.

xvn MOLECULAR WEIGHTS OF SUBSTANCES IN SOLUTION 309

several hours in copper sulphate solution. The two salts diffuse
through the porous wall, meeting somewhere inside, and producing
a coherent film of copper ferrocyanide in the wall of the pot.
The latter is now washed out, filled up with a solution, say of sugar
in water, and fitted with a manometer cemented into the open top as
shown in Fig. 172. When the pot is plunged hi to water, there is a
gradual rise of pressure in the manometer, until a steady value
is finally reached. This is called the osmotic pressure of the
solution.

The preparation of a good semipermeable pot is a matter of no little
difficulty ; most of tne results are failures, and many precautions
must be taken which cannot be described here. Better results are
obtained by driving the ions, Cu" and FeCy6'"', by electrolysis into
the pot.

The laws of osmotic pressure. — The osmotic pressures of solutions
of moderate concentrations are very considerable, as will be seen
from the results on page 310 of Morse and Frazer (1905-1913) for cane-
sugar. The concentrations are in gm. mol. (C12H-22^n — 342) per
kgm. of water.

From these figures some important results are easily deduced. We
shall at first consider dilute solutions, less than 0-5 molar (i.e., less
than 0-5 gm. mol. per litre).

If we divide the pressures at 0° by the concentrations (omitting
the anomalous figure for 0-1 molar) we find :

Concentration, C = 0-2 0-3 04 0-5 gm. mol./kgm. H20.
Pressure, P = 4-722 7-085 9442 11-895 atm.

Ratio Pf C =23-6 23-6 23-6 23-8.

The ratio is practically constant, hence the osmotic pressure, at a
constant temperature, is proportional to the concentration. This is the
exact analogue of Boyle's law for gaseous pressures..

If we next consider the osmotic pressures at various teperatures,
taken on the absolute scale, for a fixed concentration, we find,
for 0-2 molar :

Abs. temp, T 273 278 283 288 293 298°
Pressure, P 4-772 4-818 4-893 4-985 5-064 5-148 atm.
Ratio PfT 0-0175 0-0173 0-0173 0-0173 0-0173 0 0173

The ratio is constant, hence the osmotic pressure, for a given
concentration, is proportional to the absolute temperature. This is
the exact analogue of Gay-Lussac's law for gaseous pressures.

310

INORGANIC CHEMISTRY

CHAP.

Concen-
tration.

Temperature.

0° '

5°

10°

15°

20°

25°

0-1

(2-462)

2-452

2-498

2-541

2-590

2-634

0-2

4-722

4-818

4-893

4-985

5-064

5-148

0-3

7-085

7-198

7-334

7-476

7-605

7-729

0-4

9-442

9-608

9-790

9-949

10-137

10-296

0-5

11-895

12-100

12-297

12-549

12-748

12-943

0-6

14-381

14-605

14-855

15-144

15-388

15-624

0-7

16-886

17-206

17-503

17-815

18-128

18-434

0-8

19-476

19-822

20-161

20-535

20-905

21-252

0-9

22-118

22-478

22-884

23-305

23-717

24-126

1-0

24-825

25-283

25-693

26-189

26-638

27-053

Concen-
tration.

Temperature.

30°

40°

50°

60°

70°

80°

0-1

2-474

2-560

2-635

2-717

—

0-2

—
«

5-044

5-163

5-278

5-437

—

0-3

7-647

7-844

7-974

8-140

—

0-4

10-295

10-599

10-724

10-866

—

0-5

12-978

13-355

13-504

13-666

13-991

0-6

15-713

16-146

16-319

16-535

16-820

—

0-7

18-499

18-932

19-202

19-404

19-568

-p

0-8

21-375

21-803

22-116

22-327

22-567

23-062

0-9

24-226

24-735

25-123

25-266

25-562

25-919

1-0

27-223

27-701

28-213

28-367

28-624

28-818

xvn MOLECULAR WEIGHTS OF SUBSTANCES IN SOLUTION 311
Thus, dilute solutions obey the two gas laws when " osmotic
pressure " is substituted for " gas pressure." A much more striking
result may, however, still be brought to light. If 1 gm. mol. of an
ideal gas is confined in a space of 22-24 litres at 0° it will exert a
pressure of 1 atm. Boyle's law then shows that if the volume
is now reduced to 1 litre, the resulting pressure will be 22-2 atm.
The gas has then unit concentration. The ratio P/C in the table,
i.e., the pressure for unit concentration, is, however, practically
constant, and equal to 23-6. This is, within a few per cent., equal to
22-2, hence the osmotic pressure of a solution is equal to the gas
pressure which the solute would exert if all the solvent were removed,
and the dissolved substance were left in the space in the condition of
an ideal gas. Solutions therefore obey Avogadro's law.

Van't Hoff, to whom these results are due (on the basis of the older
and less accurate experiments of Pfeffer), summarised them in the
statement that dissolved substances obey the gas laws. This is known
as Van*t Hoff's gaseous theory of solution ; more accurate experi-
ments, such as those quoted above, show that it is only approxi-
mately true, but it is probable that the laws are exact only in the
limiting case of extreme dilution, just as the gas laws are exact only
at infinitely small pressures. The gaseous theory of solution is the
basis of modern physical chemistry ; its consequences have had a
most remarkable influence on the progress of the whole science
during the last thirty years. The accumulated evidence leaves no
vestige of doubt as to its truth as a broad generalisation, and the
deduction of the laws of dilute solutions from thermodynamics
strengthens this conclusion.

The Brownian movement. — An obvious step from the gaseous
theory of solution is to identify osmotic pressure with molecular
bombardment by the dissolved substance. Boltzmann was able to
show, on the assumption that the solute molecules had the same
mean kinetic energy as gas molecules, that the laws of osmotic
pressure followed from the kinetic theory. This would imply that
the molecular pressure pre-existed in the solution before the latter is
separated from the solvent by the semipermeable wall, and
that the function of the latter is merely to make the pressure evident.
The idea met with great opposition, and gradually dropped out of
sight, until it was revived, and put on the basis of an experimental
fact, by the fascinating researches of Jean Perrin, professor at the
Sorbonne. (" Les Atomes," 5th edit., 1914.)

If an aqueous suspension of gamboge, a gum-resin familiar to
painters in water-colour, is examined under the microscope, the
particles are seen to be in motion. Each particle performs little

312 INORGANIC CHEMISTRY CHAP.

excursions in an apparently erratic manner, moving in a zigzag
path. This motion was first observed with grains of pollen sus-
pended in water by the botanist Robert Brown in 1 827 ; it is shown
by all suspensions, if the particles are sufficiently small, and is
known as the Brownian movement.

The cause of the Brownian movement was ascribed to molecular
bombardment of the suspended particles, by the molecules of the

VAN'T HOFF.

liquid, by C. Wiener in 1863. This was confirmed by Svedberg
in 1906 ; he found that the length of the path described agrees with
that calculated from the kinetic theory by Einstein (1905), and
by Smoluchowski (1906).

Perrin found that if the emulsion of gamboge was allowed to settle,
the particles did not fall flat to the bottom of the vessel, but re-
mained as a minute haze, extending only over a fraction of a milli-
metre, exhibiting the Brownian movement, and diminishing rapidly

Slide

— Couer glass
Emulsion

xvii MOLECULAR WEIGHTS OF SUBSTANCES IN SOLUTION 313

in density with the height. This was exactly analogous to the
diminution in density of the atmosphere ; in the latter, on account of
the small weight of the gaseous molecules, a height of some hundreds
of miles is necessary to get the same gradation in density as is
evident in less than a millimetre with the comparatively massive
gamboge particles. The gamboge particles and the gaseous mole-
cules are equally supported against the action of gravity by their
kinetic energies. By counting the numbers of particles at different
heights under the microscope (Fig. 173) it was possible to find the
law of distribution at different heights.

If n and n' are the numbers of gamboge particles per c.c. at two
heights h cm. apart, then, if the " solution " obeys the gas laws,
the osmotic pressures, p and p', are in the ratio of n to n'. The
ratio p/p, however, will be connected with the height h by the well-
known logarithmic formula giving the diminution of barometric
pressure with the height. The distance h required to produce a
given fall of pressure is inversely proportional
to the density, or molecular weight, of the
gas. To halve the density (or pressure) in an
oxygen atmosphere, a vertical ascent of 5 kilo-
metres is required ; in hydrogen, with lighter
molecules, the ascent is 5 X 16 = 80 km., whilst
with carbon dioxide, with heavier molecules, it is
only 5 X 16/22 = 3-63 km. The " molecular
weight " of the gamboge particles could thus
be calculated from the height in which the
number per c.c. is halved. The weight of each
particle of gamboge was found by counting" the
number per c.c., and finding the total weight
per c.c. The number of particles required to make up the
molecular weight could thus be calculated. This was found to be
N = 6x 1023, which is the same as the value of Avogadro's constant
for a gas.

The suspended particles in the gamboge emulsion, therefore,
obey the gas laws. It seems very probable that the particles in
true solutions, which are much more closely similar to those of
gases, should also obey the gas laws, and that osmotic pressure is
caused by molecular bombardment. A partition allowing only
water molecules to pass through, and arresting gamboge particles,
would be subjected to a feeble bombardment by the latter, and
experience a small osmotic pressure. In the case of true solutions,
the number of molecules in a given space is much larger and the
pressure is correspondingly greater.

By examining the Brownian movement of the suspended particles
in tobacco-smoke, de Broglie found N = 643 X 1023.

Liquid diffusion. — Liquid diffusion, mentioned on p. 258 as evidence

Microscope

PIG. 173.— Perrin's

Experiment with

Gamboge Emulsion.

314

INORGANIC CHEMISTRY

CHAP.

of molecular motion, was investigated by Graham (1850-62).
He placed small bottles, containing solutions of various substances,
in large jars of water (Fig. 174), and determined
by analysis the amount of substance diffusing
into the water in a given time.

By using apparatus of the same dimensions, he
was able to obtain comparative results, and found
that the rates of diffusion differed considerably.
Acids and salts diffused fairly quickly, whereas
glue, starch, and albumin diffused only very
slowly. The rapidly diffusing substances were
(except acids) all crystalline in the solid state, and
were called crystalloids by Graham. Gum and
albumin, however, form amorphous solid masses
resembling glue, and were called colloids (Greek
kollos, glue). The differences were so great that
Graham considered himself justified in differenti-
ating between " two worlds of matter, the crystalloid and the
colloid," each with characteristic properties.

FIG. 174.

Graham's Experi-
ment on Liquid
Diffusion.

Substance.

Sodium chloride

Ammonia

Alcohol.,.

Glucose

Gum arabic

Albumin

Times of equal
diffusion.
100
60
200
300
700
2100

Amounts diffus-
ing in equal times.
100

85
47
36

0-8

0-03

Dialysis. — In another set of experiments Graham placed the solu-
tion in a shallow bell-jar, the bottom of which was closed by a piece
of parchment paper or bladder (i.e., a solid ^r^r 'TTm>.
colloid) . This membrane separated the solu-
tion from pure water, in which the appara-
tus, called a dialyser (Fig. 175), was placed.
Crystalloids passed readily through the col-
loidal septum, whereas colloids were either
arrested or diffused exceedingly slowly.

By means of the dialyser a solution of a
colloid may be freed from crystalloidal im-
purities (e.g., salts). A convenient dia-
lyser consists of a parchment paper tube
(prepared by treating unglazed paper with
concentrated sulphuric acid, and washing),
bent into a U -shape, filled with the solu- FlG- l75.-Graham's Dialyser.
tion, and placed in a jar of distilled water, which is frequently

xvii MOLECULAR WEIGHTS OF SUBSTANCES IN SOLUTION 315

renewed (Fig. 176). Small " thimbles " of parchment paper, slipped
over the end of a glass tube, and fixed by a short length of rubber
tubing, may also be used. Collodion films (p. 570) are still more
efficient.

EXPT. 122. — Pour a solution of potassium iodide and starch into a
dialyser, consisting of a piece of parchment paper tied tightly over the
mouth of a bell -jar. Suspend the bell -jar with the parchment paper
dipping into distilled water in a dish. After half an hour add chlorine
water to the water in the dish. A yellow colour, due to liberated iodine,
shows that the iodide has diffused through the parchment paper, but
the starch is retained, since this would have given a blue colour with
the iodine, as may be seen by adding
chlorine water to the liquid in the bell -jar.

All the experimental data show that
the transition from crystalloids to colloids
is gradual, depending on the size of the
particles ; suspensions of gold may be
prepared which range from microscopic-
ally heterogeneous, through colloidal
solutions (ultra-microscopically hetero-
geneous), to true solutions, with in-
creasing fineness of the particles from
10~5 cm. to 10~8 cm.

The sharp differentiation between
crystalloids and colloids made by
Graham has thus not been confirmed.
Albumin may be obtained in a crystalline
form, and crystalline substances, such as
common salt, may be prepared in the
form of colloidal solutions by precipi-
tation in liquids (e.g., ether) in which they do not form true solutions.
The real factor determining whether a substance forms a colloidal
solution or a true solution is the size of the dispersed particles
(p. 8 ) ; it is more correct to speak of the colloidal state of matter than
of " colloidal substances." Even carefully filtered solutions of cane-
sugar show a slight Tyndall effect with a beam of light (p. 7),
although this is very much less than that obtained with colloidal
solutions, which contain larger particles. Lord Rayleigh showed
that the blue colour of the sky, which was formerly attributed to the
scattering of light by suspended dust, could be accounted for by the
scattering effect of the gaseous molecules of the atmosphere.

Molecular weights of colloids. — Organic colloids must have high
molecular weights ; thus, gum arabic, although possessing the
empirical formula Gl2RnOn, is acidic, and the very small amount of
base required for its neutralisation shows that its molecule is much

FIG. 176. — Tubular Dialyser.

316 INORGANIC CHEMISTRY CHAP.

more complex . (C^HuOnV By the method of depression of
freezing point applied to other colloidal solutions, high molecular
weights have also been found: starch, 25,000; tannin, 1100;
silicic acid (p. 745), 49,000 ; rubber (in benzene), 6500. The slow-
ness of diffusion and dialysis is readily understood when one con-
siders that with such enormous molecules (often ultra-microscopic-
ally visible) the molecular movement must be very slow, since the-
square of the velocity is inversely proportional to the molecular
weight. The osmotic pressures of colloidal solutions are, as would be
expected from the large molecular weights, very small, but they
appear to be definite. Pfeffer obtained the following values with
1 per cent, aqueous solutions :

Pressure Molecular

cm. Hg. weight.
Potassium nitrate 178

Cane-sugar ... 47 342

Dextrin 16-5 975

Gum arabic ... 7'2 2230 [(C12HnOu)7 = 2317]

Since the molecular weights are inversely proportional to the
osmotic pressures (except in the case of potassium nitrate, which is
an electrolyte and is abnormal, as will be shown later), the figures
in the third column may be calculated from the osmotic pressures
and the molecular weight of cane-sugar = 342.

Liinebarger (1892), using a parchment-paper membrane, found the
molecular weight of colloidal tungstic acid, by the osmotic method,
to be 1720, which corresponds with (H2W04)7 = 1750.

Graham's suggestion that colloids as a class have high molecular
weights, and complex molecules, possibly formed by the association of
a number of crystalloid molecules (e.g., in the case of tungstic acid),
has therefore been confirmed.

The molecular weights of colloids have also been determined from
the rate of diffusion ; the latter is inversely proportional to the
square-root of the molecular weight. In this way Herzog (1908)
found the molecular weight of albumin to be 17,000 ; Sabanejeff
and Alexandroff found 13,000—14,000 by the freezing - point
method. The satisfactory agreement in this and other cases,
between results obtained by different methods, seems to indicate
that colloids possess definite molecular weights, which may, of course,
vary with the method of preparation.

Electrolytes. — In a large number of cases, the molecuiar weights
of dissolved substances are found to be the same as those deduced
from the vapour densities. When the substance is not volatile, it
often corresponds with the simplest molecular formula — e.g.,
cane-sugar, C12H22On. Solutions of organic substances in water,
alcohol, and ether usually show normal molecular weights. Raoult,

xvn MOLECULAR WEIGHTS OF SUBSTANCES IN SOLUTION 317

however, observed that many substances dissolved in benzene,
nitrobenzene, and ethylene dibromide gave depressions of freezing
point, or lowerings of vapour pressure, only half the normal, and
he explained this by the association of the solute to form double
molecules. Many such substances in fact (e.g., acetic acid) gave
abnormally high vapour densities. An abnormally small depression
is also produced when the dissolved substance crystallises out with
the solvent to form a homogeneous solid solution (p. 94) ; the
freezing point may thus even be elevated.

But when aqueous solutions of acids, bases, and salts, (i.e.,
electrolytes) were found to give molecular depressions considerably
in excess of the normal, which increased with dilution until they
approached double the normal depression in most cases, or an even
higher multiple in others, the interpretation was by no means clear.
It might indeed be supposed that all the so-called normal depressions
produced by organic solutes were really due to double molecules, and
that acids, bases, and salts are normal, but the identity of the values
of the gas constant R from measurements of gaseous density and
osmotic pressure, together with the whole body of experimental
evidence, tells unmistakably against this hypothesis. The only
other explanation possible, if we regard the laws of solution as
valid in all cases, is to suppose that the salts are dissociated in
solution. The molecules must then break up into sub-molecules,
and at high dilution the dissociation must be practically complete.
This, however, is exactly the state of affairs postulated by Arrhenius
in 1887 in his theory of electrolytic dissociation (p. 283). The sub-
molecules are the electrically charged ions : KC1 ^ K' -{- Cl', and
the increase in the number of molecules of solute so produced
accounts for the abnormally large depression of freezing point.
The ions, in fact, behave in respect to depression of freezing point
exactly like neutral molecules. The electrolytic dissociation theory,
therefore, not only gave a clear explanation of the facts of electro-
lysis, as discovered by Faraday, but cleared away in one stroke the
perplexing difficulties which had surrounded the properties of solu-
tions of electrolytes as investigated by Raoult.

Relations between different methods for the determination of
molecular weights of dissolved substances. — At first sight it would
seem that no two sets of phenomena could be less related than the
osmotic pressure and the freezing point, or vapour pressure, of
solutions. It has been well said, however, that the business of
science is to bring out unsuspected relations between phenomena,
and Newton's demonstration that the tides, and the fall of an apple
from a tree, are two expressions of an identical force operating in
Nature is only one of the many cases which could be cited in this
connection. In 1886 Van't Hoff was able to show that the osmotic
pressure, vapour pressure, and freezing point of a solution are closely

318 INORGANIC CHEMISTRY CHAP.

connected, so that if one is given the others may be calculated
without knowing anything beyond the properties of the pure solvent.
Thus, the depression of freezing point and the lowering of vapour
pressure may be calculated from the latent heat of fusion, or of
evaporation, and the freezing, or boiling, point of the pure solvent,
respectively, quite independently of the nature of the dissolved
substance. The three methods are interconnected, and necessarily
give the same results. It follows that the values of the degree of
ionisation of an electrolyte determined by all three methods must
be identical, and their agreement is, in itself, no proof of the validity
of the theory of electrolytic dissociation. On the other hand, the
conductivity (p. 291) is an entirely independent method of finding the
ionisation, and the agreement between the value so found and that
found by any or all of the other three methods, affords a very valuable
confirmation of the ionisation hypothesis.

Concentration Ionisation Ionisation

Substance. ' , ,,., from conduc- from freezing

tivity. point.

NaCl ... 0-001 98-0 98"4

0-01 93-5 90-5

01 84-1 84-1

K2SO4 ... 0-001 92-3 94-2

0-005 85-8 88-7

005 70-1 726

HC1 ... 0002 100-0 98-4

0-01 98-9 95-8

O'l 93-9 88-6

SUMMARY OF CHAPTER XVII

The freezing point of a solvent is lowered by a dissolved substance,
and the depression, D, is proportional to the amount of substance, m,
in 1 kgm. of solvent. The molecular lowering, A, for the molecular
weight, M, in 1 kgm. of solvent, is constant for all substances (except
electrolytes, and associated substances) in a given solvent, when the
solution is dilute. Thus, it follows that m : M : : D : A, or M = m A/Z).

The boiling point of a solvent is raised by a dissolved substance, and
the same laws hold as for the freezing point : M = mE/i>, where E
is the molecular elevation of boiling point.

The vapour pressure of a liquid is lowered by a dissolved substance.
If n gm. mol. of the substance are dissolved in N gm. mol. of solvent,
and if /0, / are the vapour pressures of the pure solvent and solution,
respectively, then (/0 — /)//„ = n/(N + n).

The above relations enable one to determine the molecular weight
of a substance in solution.

The osmotic pressure of a dissolved substance is related to the concen-

xvii MOLECULAR WEIGHTS OF SUBSTANCES IN SOLUTION 319

tration and temperature of the solution in the same way as the pressure
of a gas.

Colloidal solutions show only small differences from the freezing and
boiling-points of the solvent, and small osmotic pressures. The colloidal
substance has, therefore, a high molecular weight.

EXERCISES ON CHAPTER XVII

1. What methods may be used to determine the molecular weight
of a substance in solution ? Describe carefully how you would find
the molecular weight of dissolved cane-sugar by any one of these
methods.

2. Describe the cases of abnormal depression of freezing point which
are met with. What explanations have been given of these results ?

3. What is meant by osmotic pressure ? How have osmotic pressures
been measured, and of what theoretical value are the results ?

4. Van't Hoff (1886) stated that "dissolved substances obey the
gas laws." On what experimental evidence is this statement based ?
Show, from the results for the osmotic pressures of dilute solutions of
cane-sugar (p. 310), that these may be summarised in the formula
PV = RT, where P = osmotic pressure, V = volume containing 1
gm. mol., and that R has the same value as for a gas (p. 149).

5. A solution of 9'21 gm. of mercuric cyanide, Hg(CN)2, dissolved
in 100 gm. of water has a vapour pressure at 100° of 755-2 mm. Find
the molecular weight of the dissolved salt. What inference may be
drawn as to the electrolytic dissociation of mercuric cyanide in water ?

6. A solution of 9-472 gm. of cadmium iodide, CdI2, in 44-69 gm. of
water boils at 100-303°. What is the molecular weight of dissolved
cadmium iodide ?

7. A solution of lithium chloride containing 4-13 gm. per litre of
water freezes at — 0-343°. What is the degree of ionisation ? The
limiting equivalent conductivity of lithium chloride (p. 291) is 98-9:
what is the conductivity of the above solution ?

8. What is the Brownian movement ? Give a short account of
Perrin's researches on the phenomenon, and point out the importance
of the results to the theory of solutions.

9. In what respects do crystalloids differ from colloids ? To what
extent do you consider that Graham's sharp differentiation of the two
as " separate worlds of matter " is justified ?

10. A solution of 24-67 gm. of colloidal tungstic acid per litre gave an
osmotic pressure of 25-2 cm. of mercury at 17°. Find the molecular
weight. What formula does this indicate (tungstic acid is H2WO4) ?

11. Explain what is meant by dialysis : how would you separate a
mixture of common salt and albumin ?

CHAPTER XVIII

OZONE

The formation of ozone. — Van Marurn in 1785 noticed that the air
in the vicinity of an electrical machine in active operation acquired
a peculiar smell, and% tarnished mercury. Cruickshank in 1801
observed the same smell in electrolytic oxygen, but the fact that the
odour was due to a peculiar gas was only recognised fii 1840 by
Schonbein, who gave the substance the name ozone (Greek ozo, I
smell). He found that it is also produced by the slow oxidation of
phosphorus in moist air, and is capable of liberating iodine from
potassium iodide.

EXPT. 123. — Place a few sticks of freshly scraped phosphorus in a
stoppered bottle with a little water. When the fumes have subsided,
introduce a piece of paper dipped into a solution of potassium iodide
and starch (" starch-iodide paper "). This is at once turned blue. The
peculiar smell of the gas is also noticeable. The ozonisation is most
pronounced at 24° ; below 6° no action occurs, except under reduced
pressure. A greenish, phosphorescent light, which can be seen in the
dark, accompanies the formation of ozone.

Ozone is said to occur in traces in country, especially sea, air, but
many of the effects attributed to ozone are doubtless caused by
hydrogen peroxide, or oxides of nitrogen. There is some spectro-
scopic evidence for the existence of ozone in the upper atmosphere,
where it may be formed by the action of ultra-violet light on oxygen.
It has been stated that the maximum amount of ozone in the air
never exceeds 1 in 450,000. The evaporation of salt-water in the
form of spray is said to produce the ozone of sea air. If present in
larger amounts than 1 in 20,000, ozone in air has an irritant action
on the mucous membrane, and is poisonous.

Ozone is produced, apart from the action of the electrical discharge,
and the slow oxidation (autoxidation) of phosphorus in air, in many
other reactions, in most cases in small amounts.

It is contained in electrolytic oxygen, and in the oxygen evolved
by the action of fluorine on water, by the action of concentrated

320

CH. XVIII

OZONE

321

sulphuric acid on barium peroxide, potassium permanganate, and
potassium dichromate. It is produced by passing oxygen over
heated manganese dioxide, by the action of radium salts on oxygen,
and by heating ammonium persulphate with nitric acid.

EXPT. 124. — -Warm a little potassium dichromate with concentrated
sulphuric acid, and test the gas with
KI- starch paper.

Ozone is formed in traces in flames of
burning hydrogen, or coal-gas, but not by
the combustion of carbon or carbon mon-
oxide. It was previously supposed to be
formed by the slow combustion of ether
vapour, but the substance produced is
probably hydrogen peroxide. Ozone is
given off on heating crystalline periodic
acid, and by exposing oxygen to ultra-
violet light, or radium emanation.

In all cases, ozone is obtained mixed
with oxygen in varying amounts : the
product is ozonised oxygen (or ozonised air).

The preparation of ozone. — The most
convenient method of preparing ozonised
oxygen is by the action of an electric
brush-discharge on oxygen, preferably dry.
Many types of apparatus are used for
this purpose, but they are all very similar in principle. One of the
most useful is probably that of Brodie (1872) (Fig. 177).

EXPT. 1 25 — The oxygen is passed slowly through the annular space be-
tween two glass tubes, the inner tube, Z>, being filled with concentrated
sulphuric acid or copper sulphate solution, and the whole apparatus
placed in a jar, E, of the same liquid. The two wires from a good.Ruhm-

FlG. 177. — Brodie's Ozoniser.

FIG. 178. — Joints for Ozone Apparatus.

korff coil dip into the two liquids, which form electrodes, and at the
same time serve to cool the apparatus. A bluish-violet glow is seen in
the glass surfaces, accompanied by a hissing noise ; there should be very
few sparks, as these destroy ozone. The gas is conducted away through
glass tubes with ground-glass joints, or joints made with paraffin wax
or ordinary corks (Fig. 178). Rubber is very quickly destroyed by
ozone, dry cork is more resistant. Air may be used instead of oxygen,

322

INORGANIC CHEMISTRY

CHAP.

but less ozone is obtained, and nitrogen pentoxide may then be present
in the gas. (The original ozoniser of Siemens [1858] (Fig. 179) consists
of two concentric glass tubes, the outer covered, and the inner lined,
with tinfoil, but the type just described is superior in many ways.)

f Metal strip]

FIG. 179. — Siemens' Ozoniser.

By cooling the oxygen to 0°, using a powerful coil, and avoiding sparks,
as much as 25 per cent, of the oxygen may be converted into ozone ;
usually the yield is much less.

EXPT. 126. — Ozonised oxygen is formed by the electrolysis of sul-
phuric acid (sp. gr. 1-1). The apparatus is shown in Fig. 180. A very
good yield is obtained with a heavy current and an anode (positive

^ — -x

electrode) composed of a narrow platinum
tube coated with glass, having a narrow line
of metal exposed, and cooled by a stream of
calcium chloride solution at — 14° passing
through.

Ozone is formed in fairly large quantities
when oxygen (or air) is exposed to ultra-
violet light. If a quartz mercury lamp
is operated under a glass bell- jar for a few
minutes, the air in the jar smells strongly of
ozone. Liquid oxygen exposed to ultra-violet
light becomes dark blue in colour, owing
to the production of liquid ozone (p. 328).
The composition of ozone. — Schonbein found that if ozonised
oxygen is passed through a glass tube heated to 400°, it loses its
smell and action on Kl-starch paper, and the gas then appears to
be ordinary oxygen.

EXPT. 127. — Attach a piece of hard glass tube by a cork joint to
the ozoniser, and heat the tube with a Bunsen flame. The issuing gas
no longer acts on Kl-starch paper.

Marignac and de la Rive (1848), and Shenstone and Baker (1908),
found that pure dry oxygen can be ozonised by an electric discharge.
Briner a,nd Durand (1908) converted a confined volume of oxygen

FIG. 180. — Ozone from Sul-
phuric Acid by Electrolysis.

XVIII

OZONE

323

completely into liquid ozone by the silent discharge in a tube of
dry oxygen, cooled in liquid air. Thus, ozone is merely a modification of
oxygen.

This conclusion was also reached by Andrews (1856), who dried
electrolytic oxygen by means of sulphuric acid, and then passed it

FIG. 181. — Andrews' Experiments on Ozone.

through two bulb-tubes (Fig. 181) containing potassium iodide
solution, and concentrated sulphuric acid, respectively. The in-
crease in weight of the two bulbs was exactly
equal to the oxygen equivalent (O = I2) of
the iodine liberated. The iodide bulb was
then replaced by a glass tube heated to 400°.
The weight of the sulphuric acid bulb remained
constant, showing that the gas contained no
hydrogen. Andrews also found that ozone
prepared in different ways (electric discharge,
electrolysis, autoxidation of phosphorus) has
the same properties.

The formula of ozone. — If ozone is a modi-
fication of oxygen, it must have the formula
Ow. Andrews and Tait (1860) first attempted
to find the formula of ozone. They filled a
tube, A (Fig. 182). with dry oxygen, which
communicated with a sulphuric acid man-
ometer, B. Sulphuric acid is without action
on ozone. On sparking the oxygen, a maximum
contraction of one- twelfth was observed. When
the tube was heated to 300°, the original
volume was restored. A glass bulb of mercury
broken inside the tube by means of a short
length of glass rod which could be shaken on it, was converted
into a black powder, and the original volume of gas was again
recovered. A bulb of potassium iodide solution broken in the

FIG. 182. — Andrews and
Tait's Experiments on
Ozone.

324 INORGANIC CHEMISTRY CHAP.

gas produced iodine, but in this case the volume of the gas
remained unchanged, although it no longer expanded after heating
to 300°, and was therefore completely converted into oxygen.

One possible explanation of the constancy of volume of the gas
when the ozone is destroyed by potassium iodide, is that the ozone
is distributed in the gas in the form of a finely-divided solid, occupy-
ing practically no volume. A more rational explanation is that
at the same moment as one portion of ozone reacts with the iodide
another portion changes into ordinary oxygen, the expansion due
to the second change being exactly equal to the contraction due to
the first. In any case, ozone is apparently denser than oxygen.

Odling, in 1861, pointed out that the reactions could be explained
on the assumption that the formula of ozone is O3 :

2KI + 03 (1 vol.) + H20 = 2KOH + 02 (1 vol.) + I2.

The formula O2+n will obviously give the same result, but Go
is the simplest, and there were no experiments pointing to a more
complicated formula.

Odling's formula was confirmed by Soret in 1866-8 by two sets
of experiments.* Soret pointed out that oxidisable bodies which
destroy ozone without change of volume, such as those used by
Andrews and Tait, give no indication of the real density of ozone.
Thus, suppose that 100 vols. of oxygen after electrisation
contract to 90 vols. Assume that 100 vols. contain 100 O2 mole-
cules, then the contracted gas must contain 90 molecules of
(O2 -f- ozone).

This change of volume can be explained by numerous formulae
for ozone, since the only condition to be satisfied is that the 90
volumes, after heating, shall expand again to 100 volumes. This
is the case, for example, with the following formulae :

70O2 70O2 80O2 80O2

20O3 30O2 10O4 OO2 O22 11O

90 100 90 100 90 100

In order to find the relative volume of ozone in the mixture,
some solvent or absorbent is evidently necessary which takes up
the whole of the ozone without liberating oxygen (as is the case
with potassium iodide). By comparing the contraction on absorp-
tion with the expansion on heating it would then be possible to
distinguish between the above cases.

Thus, if the formula is 03, the contraction on absorption is 20,

* Eau oxi/genee et ozone, in " Classiques de la Science " (III), pub. A. Colin,
Paris, 1913.'

XVIII

OZONE

325

whilst the expansion on heating is 100 — 90 = 10. If the formula
is 04, the contraction is 10 and the expansion is 10 ; if the formula
is 022, the contraction is 1 and the expansion is 10. The formula
03 thus requires that the con-
traction on absorption shall be
double the expansion on heating.

Soret found that suitable
absorbents for ozone were
certain essential oils, such as
oil of cinnamon and oil of
turpentine. He took two flasks,
of 250 c.c. capacity, with
graduated necks, filled with
ozonised oxygen and inverted
over water (Fig. 183). In one
flask the ozone was absorbed
by turpentine, when dense
white fumes were produced ; FIG. 183.— Soret's First Experiments on Ozone,
in the other it was decom-
posed by heating the flask by a flame. The contraction in the
first flask was found to be almost exactly double trie expansion

(after the gas had cooled) in the
second. Thus, Odling's formula,
O3, was confirmed.

EXPT. 128. — The apparatus shown
in Fig. 184, devised by Newth (1896),
may be used for this experiment. It
consists of two concentric glass tubes,
the inner tube fitted into the outer by
a ground-glass stopper. The inner
tube, and the glass jar in which the
apparatus is placed, contain dilute
sulphuric acid, and the two wires from
the coil dip into the liquids as shown.
By means of projections from the
inner and outer tubes a thin glass
tube containing oil of turpentine or
oil of cinnamon is held in position in
the annular space between them. A
current of oxygen is passed through
the apparatus, and the stopcocks are
closed. The three-way stopcock is
turned so as to put the manometer, containing concentrated sulphuric
acid coloured with indigo, in communication with the apparatus, and
the oxygen is ozonised. The contraction, after cooling, is read off on

FIG. 184. — Absorption of Ozone by
Turpentine.

326 INORGANIC CHEMISTRY CHAP.

the gauge. The inner tube is then twisted, so as to break the tube of
oil of cinnamon, and after absorption has occurred, the further con-
traction is read off. It will be found that the contraction on absorption
is double the contraction on ozonisation, i.e., double the expansion
which would have occurred on decomposing the ozone by heat.

The density of ozone. — If pure ozone could be obtained, a deter-
mination of its relative density would allow us to confirm the formula
O3. But pure gaseous ozone has never been prepared, so that a
different method has to be used. Soret, in a second research
(1868), made use of Graham's law of diffusion (p. 191). If we com-
pare the relative rates of diffusion of carbon dioxide (C02, density
22), ozone, and chlorine (C12, density 35-2) into another gas, say
oxygen, then if ozone has the formula O3 (density 24) it should
diffuse rather more slowly than carbon dioxide, but more rapidly
than chlorine. The relative rates of diffusion are inversely pro-
portional to the square roots of the densities :

Rate of diffusion of CO2 _ \/24 Bate of diffusion of C12
Rate of diffusion of O3 ~~ /22 ' ^ate °f diffusion of O3

In order to get over the difficulty of the dilution of ozone with
oxygen, Soret measured the relative diffusion, v/V, of each gas
mixed; with oxygen, where v is the volume of gas diffusing and
F the total volume present in the original mixture. The rate of
diffusion of the oxygen in both directions was the same in all cases ;
the rates of diffusion of the other gases were proportional to the
numbers of molecules present in a given volume (measured by F),
and inversely proportional to the square roots of the densities.
The ratios v/V were therefore inversely proportional to the square
roots of the densities of the diffusing gases.

The apparatus is shown in Fig. 185. It consisted of three glass
tubes, B, B', and (7, placed over sulphuric acid in E, and separated
by sliding glass plates with holes, as shown, so that the tubes could
be put in communication or separated. B was in every case filled
with pure oxygen. B was first full of acid, and the mixture of
one of the gases with oxygen, prepared in C in the proper propor-
tions, was transferred to B by sliding the glass partition, o. The
glass plates between B and B' had perforations, which could be
brought between the two cylinders by sliding the plate o'. Diffusion
from B to B' was allowed to go on for forty-five minutes, when the
plate o was slid back and the cylinders were again isolated. The
gas in B' could then be driven out into a solution of baryta, when
carbon dioxide was diffused, or potassium iodide, for chlorine or
ozone. The ratio of the ozone in the original gas and in the gas in
B' was determined from the ratio of the amounts of iodine liberated
by equal volumes of the gases. If u, u are the amounts of iodine

XVIII

OZONE

327

liberated by the gas in B', and that remaining in B, respectively,
then v IV = u/(u + u'). The relative rates of diffusion were thus
found to be : chlorine, 0-227 ; ozone, 0-271 ; carbon dioxide, 0-290.

The ratio of these values for ozone and chlorine is 227 /271 = 0-838.
The inverse ratio of the square roots of the densities, assuming that
ozone is O3, is A/24 /35 -2 = 0*824. The diffusion ratio for carbon
dioxide and ozone is 271/290 = 0-93,
whilst the inverse ratio of the square
roots of the densities, again assuming O3 as
the formula of ozone, is V22 /24 = 0-95.
The agreement is to 3 per cent., which is
satisfactory when it is remembered that
the ozonised oxygen contained only 5 per
cent, of ozone by volume.

In 1898 Ladenburg repeated the experi-
ments with nearly pure ozone, obtained
by the fractionation of the liquid (p. 328).
He compared the times of effusion of
equal volumes of this gas and of oxygen
in a Bunsen's effusion apparatus (p. 263),
and found 430 sees, and 367-4 sees.,
respectively. The squares of the times of
equal effusion are proportional to the densi-
ties (p. 191), hence 4302 : 367 -42 : : x : 16.
Thus, x = 22. Since the gas contained a
little oxygen, which would make the
density lower, this result is sufficiently
near the value 24, corresponding with 03,
to confirm the latter formula.

The formula O3 for ozone was, however,
completely established by a masterly
research of Sir Benjamin Brodie in 1872.
The description of this is too long to be
given here, but the results confirmed
Soret's less accurate values in every
particular. All other formulae were shown
to be excluded.

The formula O3 shows that ozone is an

FIG. 185. — Density of Ozone
by Diffusion (Soret).

allotropic modification of oxygen (p. 114).
The cause of allotropy in this case lies in
the different molecular complexities. Ordinary oxygen has the
formula O2, whilst ozone contains three atoms of oxygen in the
molecule. O3. Both substances contain the same element, oxygen.
Ozone is called a polymer of oxygen ; the property of a substance
existing in two or more forms of different molecular weights is
called polymerism.

328 INORGANIC CHEMISTRY CHAP.

Stability of ozone. — Ozone contains considerably more energy
(p. 387) than the oxygen gas from which it is produced : it is an
endothermic substance : 302 = 203 — 2 X 34 kgm. cal. Like other
endothermic substances (p. 390), it is stable at high temperatures. If
oxygen is strongly heated, some ozone is produced : 302— 2O3. Thus,
at "6640°, there would, according to Nernst's calculations, be 10 per
cent, of ozone in the equilibrium mixture. As the temperature
falls the ozone rapidly decomposes, but if the hot gas is suddenly
chilled, the rate of decomposition becomes so slow that the decom-
position of the ozone, which is then really less stable than at
higher temperatures, is arrested. Ozone is therefore produced in
hydrogen or acetylene flames, or when a platinum wire or Nernst
filament is strongly heated by an electric current, under liquid
oxygen. -This indicates a possible method for the manufacture of
ozone.

The properties of ozone. — Ozonised oxygen, as usually prepared,
does not contain more than 15 per cent, by volume of ozone. If
the gas is cooled by passing it through a tube immersed in liquid
oxygen, deep-blue liquid ozone, b. pt. — 119°, condenses. On careful
evaporation this gives a deep-blue gas, containing about 84 per
cent, of ozone. The liquid is fairly stable below its boiling point,
and may be distilled in the entire absence of dust or organic matter,
the least trace of which, however, brings about its explosive decom-
position. The gas is very unstable, exploding if warmed, or brought
in contact with organic matter.

The decomposition of ozone in admixture with oxygen is slow
at low temperatures : it is almost instantaneous at 300°, and takes
place according to the equation 203 = 3O2. It is accompanied by
phosphorescence. Moisture slowly accelerates the decomposition :
reduced pressure, chlorine, oxides of nitrogen, and phosphorus
pentoxide, rapidly accelerate it.

Ozone is more soluble than oxygen in water. It is more soluble
in glacial acetic acid, or carbon tetrachloride, than in water, forming
blue solutions. It produces a remarkable effect on mercury : the
meniscus of the latter is destroyed, and the metal adheres to
glass in the form of a 'mirror. On shaking with water, the mercury
recovers its original form. This reaction, which may be due to
superficial oxidation, is very sensitive.

EXPT. 129. — Pass ozonised oxygen into a clean flask containing
a little mercury, and shake the flask. The mercury adheres to the
sides of the flask in the form of a mirror.

Ozone is decomposed catalytically in contact with metallic silver,
platinum, and palladium, and with oxides of manganese, cobalt,
iron, lead, and silver. In the case of silver, the metal, if warm,
is blackened, and an oxide is probably alternately formed and

xvm OZONE 329

reduced : 2Ag + O3 = Ag2O + O2 ; Ag2O + O3 = 2Ag -f 202. The
gas is decomposed by shaking it with powdered glass.

Barium peroxide, and hydrogen peroxide, react with ozone :

BaO • | Q -f O | • O2 = BaO -[- 2O2, but the gas has no action on

chromic acid or potassium permanganate (cf. H2O2). Sulphur
dioxide is oxidised to the trioxide, the ozone being completely
absorbed (Brodie) : 3SO2 + O3 = 3SO3. This is one of the few
reactions in which the ozone molecule oxidises as a whole.

Ozone is a powerful oxidising agent : it bleaches indigo solution,
and vegetable colours, and converts moist sulphur, phosphorus, and
arsenic into their highest oxy-acids. It liberates halogens from their
hydracids : SHI + 203 = 4H2O -f O2 + 4I2. Ammonia is oxidised
to white fumes of ammonium nitrite and nitrate ; a solution of
potassium ferrocyanide is oxidised to ferricyanide :

2K4FeC6N6 + H2O + O3 = 2K3FeC6N6 + 2KOH + 02.

The liberation of iodine from potassium iodide constitutes a very
delicate test for ozone, although iodine is liberated by other oxidis-
ing agents (e.g., H2O2), by chlorine and bromine, and by higher
oxides of nitrogen. The reaction with ozone is : O3 + 2KI -f H20
= O2 -f- I2 -f- 2KOH ; it occurs in a neutral solution, which then
becomes alkaline. Moist iodine is oxidised to iodic acid, HIO3 :
I2 + 503 -f- H2O = 2HIO3 -f 5O2. Dry iodine is converted into
a greenish powder, supposed to be I4O9, without change of volume
of the gas : 2I2 -f 9O3 = I409 -f- 902. An alkaline solution of
potassium iodide is oxidised to iodate (KI08), and period ate (KIO4).
Carbon compounds containing double linkages (p. 250) add on
ozone to form unstable ozonides, which are decomposed by water
with the formation of hydrogen peroxide :

H2C:CH2 + 03 = CH2 CH2

Ethylene | |

Ethylene ozonide

H<>C C/H2 ""^ H^C CH2

-O-

! H00

O GO

Formaldehyde

I! 4- II . + H202

This reaction, in which compounds probably containing a chain
of three oxygen atoms, — O — O — O — , are produced, points to the
structural formula

O O

\0/
for ozone. The readiness with which the additional atom of

330 INORGANIC CHEMISTRY CHAP.

oxygen is split off, leaving a residue of oxygen gas, 02, led to the
assumption that one atom in the ozone molecule was quadrivalent :

O=0=O. The formula

has also been proposed, but the simpler formula

is now regarded as the most probable (cf. the formula of hydrogen
peroxide, p. 341).

The existence of closed rings of oxygen atoms containing four or

o— o

more atoms: | , is not impossible, and Harries (1911) thought

that oxozone, O4, was contained in ozonised oxygen, and that the later
fractions of the gas obtained by the fractionation of liquid ozonised
oxygen liberated more iodine from, potassium iodide than corresponded
with the density, and probably contained O4. This has not been con-
firmed.

An aqueous solution of ozone reddens litmus paper before bleach-

HOviv
ing it, and has been supposed to contain ozonic acid, yO = 0.

By the action of ozone on solid caustic potash a yellow peroxide,
K204, is obtained. This is regarded by Baeyer and Villiger as
potassium ozonate, but on acidification it does not give ozone, but
only oxygen and traces of hydrogen peroxide, H2O2.

Tests for ozone. — The difficulty of detecting ozone, when it is not
present in sufficient concentration to exhibit its characteristic
smell (1 volume in 500, 000),* is that hydrogen peroxide vapour
(H2O2), and some oxides of nitrogen (N263,NO2,N204)? also liberate
iodine from potassium iodide. Papers soaked in a solution of
potassium iodide and starch are therefore of little value in the detec-
tion of ozone in the air, since the preceding compounds, and also
chlorine (which liberates iodine from an iodide : 2KI -f- C12 =
2K01 -f- I2), may be present. The lower oxides of nitrogen cannot
exist in a gas simultaneously with excess of ozone, as they are at once
oxidised to the pentoxide, N2O5, which forms nitric acid with
moisture.

Test papers soaked in an alcoholic solution of tetra methyl base
(tetramethyldiaminodiphenylmethane) are turned violet by ozone,
straw-yellow by oxides of nitrogen, and deep blue by chlorine jor bromine,

XVIII

OZONE

331

but are unaffected by hydrogen peroxide. Paper impregnated with
benzidine is coloured brown by ozone, blue by oxides of nitrogen, blue
and then red by chlorine, but is not changed by hydrogen peroxide.

If one half of a piece of neutral litmus paper is moistened with
potassium iodide solution, and exposed to a gas containing ozone,
the wetted portion is turned blue, owing to liberation of alkali :
O3 + 2KI + H20 = O2 + I2 + 2KOH. Oxides of nitrogen would
not affect the wetted portion, but would turn the other half red,
owing to the formation of nitrous and nitric acids with moisture.
The iodine liberated by passing ozone through a neutral solution
of potassium iodide may be titrated, after slight acidification, with
sodium thiosulphate (p. 522), and the equivalent amount of ozone
(O3 = I2) calculated.
Another method of
estimation depends on ^
the oxidation of sodium
nitrite solution by
ozone : NaNO2 + 03==
NaN03 + 02. Hydro-
gen peroxide and
oxides of nitrogen are
first removed from the
gas by passing it
through a solution of
chromic acid. Hydro-
gen peroxide and
ozone are destroyed
by passing the gas
through manganese
dioxide, whilst oxides
of nitrogen pass on,
and will decolorise
dilute permanganate solution. The latter solution, in turn, will
absorb oxides of nitrogen, but allows ozone to pass through.
Hydrogen peroxide is detected by bubbling the gas through a
mixture of potassium ferricyanide and ferric chloride', which is
turned blue (p. 340).

Manufacture and utilisation of ozone. — Air or oxygen is ozonised
on the technical scale by exposure to brush discharges. The
Siemens and Halske ozoniser (Fig. 186) consists of a battery of glass
or porcelain tubes with internal tubes of aluminium, enclosed in an
iron tank of water. This is earthed, and serves to cool the apparatus.
The aluminium tubes are charged to a potential of 8000-10,000
volts, each battery of 6-8 tubes requiring half a kilowatt of
power. The Ozonair apparatus consists of two sheets of aluminium

FIG. 186. — Siemens and Halske Ozoniser.

332 INORGANIC CHEMISTRY CH. xvin

gauze separated by a plate of the insulator " micanite," several
units being enclosed in a case, and alternate plates charged and
earthed. The best production amounts to about 40-60 gin. of
ozone per kilowatt-hour of energy, at a concentration of 2 gm. of
O3 per cu. metre of air. With pure oxygen, 120-180 gm. are obtained.
The yields are only about 5 and 15 per cent, of the theoretical with
air and oxygen, respectively.

Ozonised air is used in the sterilisation of water, when it is bubbled
through the filtered water in a tall column (2 gm. of ozone per cu.
m. of water) ; for purifying air (e.g., in the Central London Tube
Railway) ; for oxidation processes (e.g.. m>-eugenol to vanillin),
and its use for other purposes is in an experimental stage. The
purification of water is its most important use : the plant supplying
Paris deals with 24,000,000 gallons daily. A small plant is in
operation at Knutsford, in Cheshire.

EXERCISES ON CHAPTER XVIII

1. In what reactions is ozone produced ? How is the substance
prepared (a) in the laboratory, (b) on the large scale ? For what
purposes is it used ?

2. What experiments would you carry out to prove the following
assertions : (a) ozone contains no element but oxygen ; (b) ozone is
a powerful oxidising agent ; (c) the formula of ozone is O3 ?

3. What is an endo thermic substance ? What do you know of the
stability of ozone at high temperatures ?

4. Describe, ' with equations, the action of ozone on (a) silver,
(b) manganese dioxide, (c) potassium iodide, (d) potassium ferrocyanide,
(e) caustic potash.

5. Give a brief accounib of the experiments which led to the adoption
of the formula On for ozone What structural formulae for the sub-
stance have been suggested ?

6. One hundred c.c. of ozonised oxygen, when shaken with turpentine,
contract to 85 c.c. What expansion will occur when 100 c.c. of the
original gas is heated to 300° ?

7. What tests would you apply to detect ozone in a gas ? Point out
what other substances might give these reactions, and say how you
would distinguish them from ozone.

CHAPTER XIX

HYDROGEN PEROXIDE

Hydrogen peroxide, H202. — Barium monoxide or baryta, BaO,
can absorb oxygen, forming a higher oxide, Ba02, called barium
peroxide. This is produced : (a) by passing a stream of oxygen
over baryta heated to dull redness : 2BaO -f- O2 ^± 2Ba02 ; (b) by
adding baryta to fused potassium chlorate, and washing out the
soluble potassium chloride from the residue with water : 3BaO -f
KC103 = 3Ba02 + KC1 (soluble). The first method is due to Gay-
Lussac and Thenard, the second to Liebig and Wohler.

If barium peroxide is added to cold dilute hydrochloric acid, no
oxygen is evolved ; the solution contains barium chloride, and a new
substance, hydrogen peroxide : Ba02 + 2HC1 — BaCl2 + H2O2.
Thenard, its discoverer (1818), called the latter oxygenated water.

The liquid acts as an oxidising agent, liberating iodine from a
neutral, or acid, solution of potassium iodide : 2KI -f- H2O2 =
2KOH + I2. From the amount of iodine liberated, the proportion
of hydrogen peroxide may be calculated.

In order to obtain a solution of hydrogen peroxide free from the
soluble barium salt, the barium peroxide must be treated with an
acid such as sulphuric, carbonic, or hydrofluosilicic (H2SiF6), which
forms an insoluble barium salt. The latter precipitates, leaving an
aqueous solution of hydrogen peroxide."

EXPT. 130. — Stir up finely powdered barium peroxide with distilled
water in a beaker, and pass a rapid stream of carbon dioxide through the
suspension. After a few minutes add a solution of potassium iodide and
starch : a blue colour is produced.

According to Merck, the above reaction should be carried out as
described, not by adding the barium peroxide in small quantities at a
time, when the particles become coated with insoluble barium carbonate.
If excess of barium peroxide is used at once, the liquid remains alkaline
until the end of the process, and decomposition is complete. An
unstable barium percarbonate, BaCO4, is first produced, which is then
decomposed by water, producing barium carbonate and hydrogen

peroxide : Ba62 + CO2 = BaCO4 ; BaCO4 + H2O = BaCO3 + H2O2.

333

334 INORGANIC CHEMISTRY CHAP.

Anhydrous oarium peroxide is not easily decomposed by dilute
sulphuric, or hydrofluosilicic, acid, on account of the formation
of a coating of insoluble compounds on the particles of peroxide.
A hydrated barium peroxide, Ba02,8H20, is however, readily decom-
posed by these acids. It is prepared as follows.

Commercial barium peroxide, containing oxides of iron and
aluminium, and silica, is finely powdered, and added a little at a time
to a cold mixture of equal volumes of water and concentrated
hydrochloric acid until the latter is neutralised. A little baryta
solution is then added, which precipitates the iron and aluminium
as hydroxides. These, together with the silica originally contained
in the barium peroxide, are filtered off, and to the filtrate is added a
saturated solution of barium hydroxide. A white, crystalline
precipitate of hydrated barium peroxide is formed, which is filtered
off, washed with cold water free from carbon dioxide, and kept
moist in a stoppered bottle : (1) BaO2 + 2HC1 = BaCL + H202 ;
(2) H202 + Ba(OH)2 + 6H20 = BaO2,8H20.

If this hydrated peroxide is treated with cold dilute sulphuric
acid (1 vol. of acid : 5 vols. of H20), or with hydrofluosilicic acid,
insoluble barium salts and a solution of hydrogen peroxide are
produced : Ba02 -f H2S04 = BaS04 + H2O0, or BaO2 + H2SiF6
= BaSiF6 + H202.

If metallic sodium contained in a nickel boat is heated in a hard
glass tube in a current of oxygen, the metal burns with a yellow
flame, and a yellow mass of sodium peroxide, Na2O2, is left.

EXPT. 131. — 6 urn a small piece of sodium in a deflagrating spoon in
a jar of dry oxygen. When the spoon is cold, dissolve the sodium per-
oxide in it by placing the spoon in water. Add dilute HC1, and a solu-
tion of KI and starch. A blue colour is produced : Na2O2 + 2HC1 =
2NaCl + H202.

Sodium peroxide is now manufactured by heating sodium in a
current of dry air, purified from carbon dioxide, and is a convenient
source of hydrogen peroxide. The calculated amount of sodium
peroxide is added, in small quantities at a time, to 20 per cent,
sulphuric acid cooled in ice : Na202 + H2S04 = Na2SO4 -f- H2O2.
Two-thirds of the sodium sulphate separates as crystals of Glauber's
salt, Na2S04,10H20 and the liquid is then decanted and distilled
in vacuo (see p. 336). Hydrogen peroxide is less volatile than
water, so that the later fractions are collected. In this way Merck
prepares a 30 per cent, solution of H202, known as perhydrol. It
is preserved in stoppered bottles covered inside with paraffin wax.

More dilute solutions of hydrogen peroxide are prepared (usually
from barium peroxide) for use in pharmacy. The strength of these
solutions is stated in terms of the volume of oxygen evolved on
heating, when the peroxide decomposes : 2H202 = 2H20 + O2.

xix HYDROGEN PEROXIDE 335

Commercial peroxide is usually " 10 volumes," or " 20 volumes,"
according as it gives off 10, or 20, times its volume of oxygen.
Merck's preparation evolves 100 times its volume of oxygen : it is
therefore sometimes (very improperly) called " 100 per cent,
peroxide."

From the equation : 2H2O2 = 2H20 -f O2, it is seen that 2 x 34
gm. of hydrogen peroxide evolve 32 gm. of oxygen, occupying 224
fitres at S.T.P. Thus each gram of peroxide evolves 353 c.c. of
O2. A 1 per cent, solution therefore evolves 3-53 times its volume
of oxygen ; " 10 vol." peroxide is therefore not quite 3 per cent,
strength.

Concentration of solutions of hydrogen peroxide. — A dilute
solution of hydrogen peroxide may be concentrated in several ways.
If it is frozen, ice separates, and the residual liquid is therefore enriched
in peroxide. It may also be concentrated by evaporation in a dish
on a water-bath : hydrogen peroxide is appreciably less volatile
than water. At a certain point, however, decomposition begins.
The solution may then be placed in a flat dish in an evacuated
desiccator containing concentrated sulphuric acid. When the
solution has reached a certain concentration of peroxide, the latter
begins to volatilise, but by working at low temperatures Thenard
was able in this way to obtain a liquid (sp. gr. 1*452) giving off
475 vols. of 02, i.e., containing 95 per cent, of H2O2.

Hydrogen peroxide is very soluble in ether, so that if an aqueous
solution is extracted with ether in a separating funnel (p. 14), most
of the peroxide passes into the ethereal layer. The latter may be
separated, and evaporated on a water-bath, when a concentrated
solution of hydrogen peroxide in water is left.

These concentrated solutions decompose very easily on heating,
or even at the ordinary temperature. They are rendered more
stable by a trace of acid. Dilute aqueous solutions are fairly
stable, especially if acidified.

More concentrated hydrogen peroxide may be obtained by distil-
lation under reduced pressure ; this method was also used by
Thenard.

Pure hydrogen peroxide. — Until 1894, hydrogen peroxide was
known only in the form of a more or less concentrated aqueous
solution. In that year Wolff enstein obtained practically pure
hydrogen peroxide by the fractional distillation of a concentrated
aqueous solution under reduced pressure. He found that, under
special conditions, hydrogen peroxide is fairly stable towards heat,
viz., when it is free from (a) all alkaline substances, (6) every
trace of heavy metal compounds, (c) all kinds of solid bodies, even
of otherwise indifferent chemical character, e.g., silica, alumina, etc.
(The sodium sulphate in Merck's method of preparation (p. 334) is
quite indifferent towards hydrogen peroxide.) By evaporating a 4 -5

336

INORGANIC CHEMISTRY

CHAP.

per cent, solution of the peroxide in a porcelain dish on a water-bath
at 75°, he concentrated it to 66-6 per cent. Some peroxide was
lost, not by decomposition, but by evaporation in the escaping
steam, since it is distinctly volatile. This solution was shaken
with ether, to precipitate alumina, and the ether evaporated from
the filtered liquid on a water-bath. The strong hydrogen peroxide
remaining was then distilled under the reduced pressure of
65 mm., and the fraction coming over between 81° and 85°
collected. It contained 90-5 per cent, of H2O2. This was again
fractionated under reduced pressure, and the fraction between 84°
and 85° collected. It contained 99*1 per cent, of H202, and was
free from all impurities.

The apparatus used for distillation under reduced pressure consists

(Fig. 187) of a distilling
flask, containing the
solution of hydrogen
peroxide, fitted with a
thermometer, and placed
on a water-bath. The
side tube of this flask
is fitted by a rubber
stopper to the inside
of a second distilling
flask, which serves as a
receiver, and is cooled
by a stream of cold
water. The side tube
of this flask communi-
cates by pressure tub-
ing with a large empty
bottle, which is con-
nected with a good
metal water pump,
working on a high-
pressure tap. A pressure gauge is connected with this bottle, and a
three-way stopcock allows air to be admitted to the apparatus when
the experiment is finished, so that the different parts may be dis-
connected, or when the receiver is changed during the operation.
In the fractionation of hydrogen peroxide there is some danger of
explosion, when the whole apparatus is shattered. This appears to
be due to some extent to the formation of an unstable ethyl peroxide,
(C2H5)202, discovered by Brodie, which is produced from the ether
remaining in the peroxide after evaporation. It is safer to begin
the experiment directly with Merck's 30 per cent, perhydrol, which
has not been treated with ether.

Pure hydrogen peroxide is a clear, syrupy liquid, colourless in

FIG. 187. — Distillation under Reduced Pressure.

XIX HYDROGEN PEROXIDE 337

small amounts, but having a bluish colour like water when in
bulk. It has an odour like that of nitric acid. It evaporates spon-
taneously in the air, boils at 84-85 °/68 mm. or 69-2°/26 mm. Its
specific gravity is 1 458 at 0°. The liquid has a strong acid reaction
to litmus. In dilute solution (1*5 per cent.), however, hydrogen
peroxide is completely neutral. By mixing the substance with
water,- and cooling in a mixture of solid carbon dioxide and ether,
the crystalline hydrates : H2O2,H20, and H202,2H20, are obtained.
The pure substance is fairly stable, and can be kept for several
weeks in the absence of sunlight, provided the glass of the bottle is
perfectly smooth. In contact with rough surfaces, or on shaking,
decomposition occurs : 2H2O2 = 2H2O -f- O2. Finely divided
metals such as gold, silver, and platinum (but not iron) bring about
explosive decomposition. Cotton- wool at once inflames. A mixture
of magnesium, or carbon, powder with a trace of manganese dioxide
at once inflames in contact with pure liquid H202.

On cooling 95-96 per cent, peroxide in solid carbon dioxide and
ether, or in methyl chloride at — 23°, it solidifies to a hard crystal-
line mass. If a little of this solid is placed in the 95 per cent, solution
cooled to — 10°, columnar prismatic crystals of pure solid hydrogen
peroxide, melting at — 2°, are obtained. These crystals explode
with a trace of platinum black ; alone, they are fairly stable.

Solutions of hydrogen peroxide readily decompose spontaneously
into water and oxygen in presence of traces of alkali. They become
much more stable in presence of traces of sulphuric or phosphoric
acids (hence the commercial peroxide is acid). The addition of
alcohol, glycerin, or barbituric acid also renders the solutions
stable. The vapour of hydrogen peroxide appears to be quite stable.

The chemical properties of hydrogen peroxide. — Hydrogen peroxide
closely resembles ozone in many respects. It is an endothermic
compound : H2-f-02 = H202 — 45-2 kgm. cal. It is therefore
unstable at the ordinary temperature, and, as in the case of ozone,
one of the oxygen atoms tends to split off, with the formation of
gaseous oxygen and water : 2H202 = 2H20 + 02 + 203-2 kgm.
cal. This decomposition is seen to be attended with a very targe
evolution of heat, much greater than that which would be evolved
in the decomposition into the elements : 2H202 = 2H2 -f 202 -f
904 kgm. cal. It is therefore the former reaction which actually
occurs.

As in the case of ozone, the endothermic hydrogen peroxide is
produced, and is stable, at high temperatures. If a hydrogen or
carbon monoxide flame is allowed to impinge on the surface of cold
water, ice, or solid carbon dioxide, hydrogen peroxide is found in
the liquid. By rapid cooling, the hydrogen peroxide formed is
chilled to a temperature at which its rate of decomposition is small
before much decomposition at intermediate temperatures can occur.

338 INORGANIC CHEMISTRY CHAP.

Traces of hydrogen peroxide are formed by the direct union of
hydrogen and oxygen when the mixed gases are passed over palladium -
black : the water formed gives the reactions of the peroxide : H2 -f O2
= H2O2. Small amounts are also formed by the action of bright sun-
light, ultra-violet light, or radium emanation on -ater containing dis-
solved oxygen : or by the action of a brush discharge on a mixture of
steam and oxygen, 2H2O + O2 = 2H2O2. It is not produced by the
spontaneous evaporation of water in air unless traces of zinc are present,
although snow is said to contain it in traces. Minute quantities of
hydrogen peroxide are formed in growing plants.

EXPT. 132. — Allow a hydrogen flame to impinge on a piece of ice. Pour
out the liquid produced, and add a little potassium iodide and starch
solution : a blue colour indicates the presence of hydrogen peroxide.

Hydrogen peroxide is an active oxidising agent, the labile oxygen
atom being easily split off, with formation of water. Arsenious
and sulphurous acids are oxidised to arsenic and sulphuric acids :
H3As03 + H202 = H3As04 + H20 ; H2SO3 + H202 = H2SO4 +
H2O. Black lead sulphide is oxidised to white lead sulphate :
PbS + 4H2O2 == PbS04 + 4H20, a reaction which is utilised in
restoring discoloured oil-paintings, in which the white-lead pigment
(basic lead carbonate) has become converted into black PbS by
atmospheric sulphuretted hydrogen. Ferrous and manganous
salts in neutral solution are converted into insoluble ferric oxide and
manganese dioxide, respectively ; from ferrous salts in acid solution
ferric salts are formed : 2FeS04 + H2O2 -f H2S04 = Fe2(S04)3
+ 2H20. This reaction may be used in the estimation of hydrogen
peroxide. Benzene, in the presence of ferrous sulphate, is oxidised
to phenol : C6H6 -f H2O2 = C6H5OH -f H2O, and hydrogen per-
oxide is used as an oxidising agent in many organic oxidations when
more energetic reagents would cause decomposition. Hydro-
gen peroxide also forms molecular compounds with many organic
substances, and with some salts : (NH4)2S04,H202, K2CO3,2H202,
Na2HPO4,H202. In these compounds it shows analogies with
water of crystallisation.

Hydrogen peroxide is a feeble acid, much weaker than carbonic
acid. With ammonia it forms directly the salt-like compounds,
NH4-O2H (ammonium hydrogen peroxide), and (NH4)202 (ammo-
nium peroxide). The compounds Na02H and Na2O2 are also known.
The oxidising action of hydrogen peroxide is used in bleaching
delicate materials (wool, silk, ivory, feathers) which would be
injured by chlorine : the solution of the peroxide is made faintly
alkaline with ammonia, or added to a 10 per cent, solution of sodium
acetate. Hydrogen peroxide bleaches hair to a golden-yellow
colour : it is called an auricome when used for this purpose. It is
also a powerful antiseptic, and as it leaves no injurious products after

xix HYDROGEN PEROXIDE 339

its action, it is largely used as a gargle, etc. Hydrogen peroxide is
used as an antichlor to remove excess of chlorine from bleached
fabrics : C12 + H2O2 = 2HC1 + O2.

Platinum black, and especially colloidal platinum (prepared by
striking electric arcs between platinum wires under distilled water),
bring about a rapid catalytic decomposition of hydrogen peroxide :
2H202 = 2H2O + O2.

EXPT. 133. — Add a little colloidal platinum to a solution of H2O2.
There is a brisk evolution of oxygen. Stirring the liquid with a glass
rod accelerates the reaction.

Liebermann (1904) considered that the platinum first absorbs
atmospheric oxygen, rendering it " active," and the activated
oxygen, probably in the atomic condition, then reacts with the

labile oxygen atom of the peroxide : H2O • ! O" +O j = H2O + O2.

Finely divided silver (see below), manganese dioxide, and other
substances also cause the catalytic decomposition of H202.

In certain reactions hydrogen peroxide appears to function as a
reducing agent. Thenard (1819) found that gold and silver oxides
are reduced by it to the metals : H2O2 -f Ag20 = H2O -f- O2 -f-
2Ag.

EXPT. 134. — Add caustic soda solution to a solution of silver nitrate :
a brown precipitate of silver oxide is formed : 2AgNO3 -j- 2NaOH =
Ag2O -f- 2NaNO3 + H2O. Add H2O2 to this : it is at once converted
into black metallic silver, with brisk evolution of oxygen. If a further
quantity of H2O2 is added, it is catalytically decomposed by the finely
divided silver.

Brodie (1850) showed that whenever hydrogen peroxide acts
as a reducing agent, it is because the labile oxygen atom can with-
draw another oxygen atom from the compound reduced, to produce
a molecule of gaseous oxygen. Thus, it reacts (rather slowly) with

ozone : O2 • ! O'+'O"! • H20 = 02 -f O2 -f H20.

EXPT. 135. — Pour hydrogen peroxide (20 vols.) into a jar of ozonised
air, replace the glass plate, and shake. After a time the odour of ozone
disappears.

A solution of potassium permanganate acidified with sulphuric
acid is readily reduced by hydrogen peroxide, with evolution of pure
oxygen (p. 162) : 2KMn04 + 3H2SO4 + 5H202 = K2SO4 +
2MnS04 + 8H2O -J- 502. This reaction may be used in the estima-
tion of H202. Persulphates are also reduced : K2S208 -f- H202 =
2KHS04 + 02.

Manganese dioxide brings about an evolution of oxygen from a
neutral solution of hydrogen peroxide, the action being apparently

z 2

340 INORGANIC CHEMISTRY CHAP.

catalytic. In an acid solution the manganese dioxide is also reduced
and a manganous salt is formed : MnO2 -f- H2O2 -f- H2S04 =
MnS04 + 2H2O -J- 02. Solutions of bleaching powder [calcium
hypochlorite, Ca(OCl)2], and sodium hypobromite also evolve
oxygen : NaOBr + H2O2 = NaBr -j- H2O -f O2. Iodine is liberated
from acidified potassium iodide : 2KI -f H2O2 + H2SO4 = K2SO4
-f 2H2O -f- I2. All these reactions are applied in the estimation
of hydrogen peroxide.

Hydrogen peroxide acts powerfully on a photographic plate. The
effects of traces of this substance have often been attributed to "rays."

An interesting case of the oxidising and reducing action of hydro-
gen peroxide was discovered by Brodie. An acid solution of potass-
ium /errocyanide is oxidised by hydrogen peroxide to potassium
/emcyanide : 2K4FeC6Ne + H2O2 = 2K3FeC6N6 + 2KOH. An
alkaline solution of potassium /emcyanide, however, is reduced to
potassium /errocyanide by hydrogen peroxide : 2K3FeC6N6 +
2KOH + H202 = 2K4FeC6N6 + 2H20 -f- 02. These changes may
be followed by the reactions with iron salts described on p. 248.

EXPT. 136. — Add some zinc to dilute sulphuric acid and ferrous
sulphate solution in a flask through which a current of coal gas is passed.
By means of a dropping funnel through the cork add a solution of
K4FeC6NG in boiled water. A nearly white precipitate is formed. Now
add ferrocyanide + H2O2 : a deep blue precipitate is produced. H2O2
evolves oxygen from an alkaline solution of K3FeC6N6.

Tests for hydrogen peroxide. — A very delicate test for hydrogen
peroxide is the liberation of iodine from potassium iodide, giving a
blue colour with starch. One part of peroxide in 25 million parts of
water may be detected by this test. Other substances, such as
ozone and nitrites, give this reaction, but hydrogen peroxide is the
only substance which liberates iodine from potassium iodide in
presence of ferrous sulphate. The reaction is : 2KI -f H202 =
2KOH + I2.

Another delicate reaction for hydrogen peroxide is the formation
of a red coloration, due to titanium peroxide, Ti03, with a solution of
titanium dioxide in dilute sulphuric acid. This solution is prepared
by heating TiO2 with twice its volume of concentrated sulphuric
acid, cooling, and diluting with ice-water.

EXPT. 137. — If hydrogen peroxide is added to a solution of potassium
dichromate acidified with dilute sulphuric acid, a brown colour is pro-
duced. If the solution is rapidly shaken with ether, the latter floats
to the surface with a beautiful blue colour. An unstable perchromic
acid is formed (p. 956), which dissolves in ether to form the blue liquid :
this decomposes after a time, with evolution of oxygen, and a green,
aqueous solution of chromic sulphate is formed in the lower layer.

xrx HYDROGEN PEROXIDE 341

Other tests are as follows : (1) guaiacol solution acidified with
sulphuric acid gives a blue colour ; (2) guaiacum tincture, with a
little blood, gives a blue colour (this is also a delicate test for blood,
and can be used in identifying blood-stains) ; (3) a mixture of
aniline and potassium chlorate, dissolved in dilute sulphuric acid,
gives a violet colour ; (4) filter-paper soaked in a solution of cobalt
naphthenate, and dried, changes from rose to olive-green with
hydrogen peroxide.

The formula of hydrogen peroxide. — The vapour-density of
hydrogen peroxide has not yet been determined, but the molecular
weight of the substance has been found from the freezing point of
its aqueous solution (Carrara, 1893) to be 34, hence its formula is
H202 (p. 302).

The constitutional formula may be written H-O-O-H, i.e., dihydroxyl,
HO -OH. This is in accordance with the instability of compounds
which contain chains of directly linked oxygen atoms.

In order to account for the instability of one oxygen atom, which
is readilv split off, Kingzett (1884) assumed the formula to be

iv M \

0 : O <^ , in which one atom of oxygen is quadrivalent. This is

in agreement with the acidic character of the peroxide : it evolves
carbon dioxide from a solution of sodium carbonate added to it
drop by drop, forming sodium peroxide : H202 -f Na2C03 =
Na202 4- H2O -f C02. (If the peroxide is addett to the carbonate,
pure oxygen is evolved by catalytic decomposition.)

Briihl (1895), from the optical properties, suggested that both the
oxygen atoms are quadrivalent : HO | OH.

By the action of hydrogen peroxide on diethyl sulphate,
(C2H5)2SO4, Baeyer and " Villiger (1900) obtained diethyl peroxide,
(C2H5)202, and ethyl hydroperoxide, C2H5H02. The former boils
at 65° and is stable ; the latter is violently explosive. By the
action of zinc and acetic acid on diethyl peroxide it is reduced
to ethyl alcohol, C,H5-OH. This agrees with the formula
02H5-0-0-C2H5 :

C2H5-0- OC2H5 = CSH6-OH + HOC2H5

i i

Kingzett 's formula, on the contrary, would require that ether,
(C2H5)2O, should be formed :

v 25 \

\0 = 0 -> >0 + H20.

f
2H

342 INORGANIC CHEMISTRY CHAP.

The formula of ethyl peroxide is therefore C2H5-0-OC2H5 ;
that of ethyl hydrogen peroxide may be written C2H5'OOH, and it
is very probable that the formula of hydrogen peroxide is H'OOH.
It is therefore a true peroxide, containing two singly linked oxygen

H-0 Na-0

atoms : | . Sodium peroxide is and barium pcr-

H-0 Na-0

/°

oxide Ba<^ | . True peroxides, which give hydrogen peroxide

with acids, differ in constitution from the dioxides of lead, man-
ganese, etc., which give only oxygen with concentrated acids,
and chlorine with concentrated hydrochloric acid. Their formulae

.0
are of the type : Pbf^ . * This is confirmed by the formation of

unstable higher chlorides on treatment with cold concentrated
hydrochloric acid (p. 224) ; e.g., MnCl4 and PbCl4. These form
complex salts, e.g., (NH4)2PbCl6, ammonium chloroplumbate.

With hot concentrated hydrochloric acid, however, barium per-
oxide evolves chlorine : Ba02 + 4HC1 = BaCl2 + C12 -f 2H20.

Autoxidation. — The formation of hydrogen peroxide during the
slow oxidation of phosphorus, oil of turpentine, and metals, by
gaseous oxygen in the presence of water, was studied by Schonbein
in 1858. He found that the oxygen is equally divided in oxidising
the substance (e.g., lead) and in forming hydrogen peroxide :
Pb + 02 -f H20 = PbO + H202.

Schonbein considered that the oxygen molecule contained a
positively and a negatively charged atom of oxygen, which he called
antozone and ozone, respectively. The antozone formed hydrogen
peroxide with water, or, if indigo solution or another oxidisable sub-
stance was present, it oxidised the latter.

It was found later that in many cases the first product is an un-
stable peroxide, which is decomposed by water with formation of
a lower oxide, and hydrogen peroxide : R -f 02 = R02 ; R02 + H2O
= RO -f- H2O2. Turpentine, for example, forms a crystalline
peroxide on standing in a loosely stoppered bottle.

According to Engler and Wild, the oxygen molecule is first opened
up to form — O — 0 — , which combines with the activator (e.g.,
turpentine) to form the unstable peroxide. In some cases these
unstable peroxides have been isolated. The bleaching and disin-
fecting properties of turpentine are due to its ability to activate
oxygen in this way.

xix HYDROGEN PEROXIDE 343

EXERCISES ON CHAPTER XIX

1. How is hydrogen peroxide prepared from (a) barium peroxide,
(6) sodium peroxide ? What are the uses of hydrogen peroxide ?

2. In what way is pure hydrogen peroxide prepared ? Give its
important properties.

3. Give examples of (a) the catalytic decomposition of hydrogen
peroxide, (6) oxidising actions of hydrogen peroxide, (c) reducing
actions of hydrogen peroxide.

4. What are the tests for hydrogen peroxide ? How is the substance
estimated ?

5. Five c.c. of a solution of hydrogen peroxide are acidified with sul-
phuric acid, and shaken with manganese dioxide. 48-3 c.c. of oxygen
(measured at S.T.P.) are evolved. How many grams of H2O2 does 1
litre of the solution contain ?

6. How has the formula of hydrogen* peroxide been established ?
Explain the difference in constitution between barium peroxide and lead
peroxide. What experimental evidence is there in support of the con-
stitutional formulae attributed to these substances ?

7. What is meant by aut oxidation ? Describe two experiments to
illustrate this process. How is it explained ?

CHAPTER XX

CHEMICAL EQUILIBRIUM, AND THE LAW OF MASS-ACTION

Chemical affinity. — In the preceding chapters chemical reactions
of various kinds have been considered, without any reference to the
possible cause of chemical chJhge. In the earlier history of chemistry
it seems to have been assumed that substances which were closely
related to one another (e.g., mercury and gold) showed the greatest
tendency to combine, hence the name affinity (from affinis, related)
was given to the cause of chemical combination. When the mutual
action of acids and alkalies was examined, it became clear that it is,
on the contrary, dissimilar substances which enter most easily into
combination, and in the electrochemical theory of Berzelius, in which
substances of opposite electrochemical character were regarded as
most prone to combination, the antithesis of the older idea found its
sharpest expression.

It was assumed by the alchemists (with the exception of Van
Helmont) that substances were destroyed on combination, so
that an acid and alkali, for instance, had nothing in common
with the salt produced from them. Boyle, in his " Sceptical
Chymist " (1661), however, remarks that : " gold may be so altered,
as to help to constitute several bodies, different from itself, and the
other ingredients ; yet it may be reduced again into the same yellow,
fixed, ponderable, and malleable gold it was, before its mixture with
them." He also observes that : " notwithstanding, the particles of
some bodies are so closely united, yet there are some which may meet
with particles of other denomination, which are disposed to be more
closely united with some of them than they are amongst themselves."
In this the elective character of chemical affinity is clearly expressed.
Mayow (1674) also held very clear views on chemical affinity. If
ammonia, he says, be added to hydrochloric acid, sal-ammoniac
is produced, in which, it is true, neither acid nor alkaline properties
are apparent. But if this is heated with potash, the ammonia is
displaced, " because the acid is capable of entering into closer
union " with potash than with ammonia. To show that an acid
is not destroyed on neutralisation, he refers to the distillation of

344

CH. xx CHEMICAL EQUILIBRIUM, LAW OF MASS-ACTION 3-J5

nitre with sulphuric acid, which displaces the nitric acid, and leaves
in the retort the same substance as is produced by the direct action
of sulphuric acid 011 potash. Nitre, on heating alone, does not lose
nitric acid, because the acid is kept down by the attraction of the
potash ; if sulphuric acid is added, the nitric acid comes off, " because
the volatile acid . . . has been expelled from the society of the
alkaline salt by the more fixed vitriolic acid." Mayow gives a
number of examples of this kind.

Similar views were held by Newton, who pointed out that potash
becomes moist in the air, whilst nitre remains dry, in consequence of
an attraction for moisture shown by the first substance, but not by
the second. Similarly, mercury precipitates silver from its solution in
nitric acid, copper in turn precipitates mercury, and iron precipitates
copper, because of the increasing attractions of these metals for the
acid. He suggested that the attractions might be electrical in
character. There is still very little known of affinity, but it appears
that Newton's speculation may be true.

Geoffrey (1718), and Bergman (1775), generalised the results, and
stated that of three substances, A, B, and C, if A has a stronger
attraction for B than C has, then A is able to decompose BC completely,
turning out C and forming AB. Tables of affinity were therefore
drawn up, giving the order in which acids, for instance, displaced
each other both in solution and in the state of fusion.

Bergman's theory of elective affinities was called into question
by Berthollet (" Researches into the Laws of Affinity," Cairo, 1799).
He pointed out that the reaction A + BC = AB -f- C does not
always proceed to completion in one direction, as it should according
to Bergman's theory. It may proceed in the opposite direction
under different conditions, and in general is not complete : "in
opposing the body A to the combination BC, the combination AC can
never take place [completely], but the body C will be divided between
the bodies A and B proportionally to the affinity and the quantity
of each."

A chemical reaction, e.g., A -{- BC = AB -j- C, may proceed only
to a certain point, because the opposed reaction : AB + C = A -j-
BC can often take place under the same conditions, and at the same
time as the direct reaction. A state of equilibrium is then reached.
wjien. the two opposing reactions balance each other, i.e., proceed
with equal speeds. This is denoted by : A -f- BC ^ AB -}-- C.

Many examples of such states have already been given. Thus,
steam is reduced by heated iron, giving hydrogen and oxide of iron :
3Fe + 4H20 -» Fe304 + 4H2 (p. 183). But under the same condi-
tions, oxide of iron is reduced by hydrogen, giving iron and steam :
Fe304 -f 4H2 -> 3Fe + 4H20. The oxygen is shared between
the iron and the hydrogen, and a state of equilibrium is set up when
the two reactions are balanced, i.e., as much steam is decomposed

346 INORGANIC CHEMISTRY CHAP.

as is produced, in a given time : 3Fe + 4H2O ^ Fe3O4 -f 4H2.
Other examples are the decomposition of barium peroxide by heat
(p. 168) : 2BaO2 ^ 2BaO + O2 ; and the dissociation of steam at
high temperatures (p. 212) : 2H2O ^ 2H2 + O2. Such reactions as
the above, which can proceed in either direction, are called reversible
reactions.

EXPT. 138. — Pour concentrated hydrochloric acid over crystals of
Glauber's salt (Na2SO4,10H2O). Filter off the white residue, wash
with a little water, dry on a porous plate, and heat with concentrated
sulphuric acid : fumes of hydrochloric acid are evolved, hence the
precipitate is sodium chloride. The two reactions are : (1) Na2SO4
-f 2HC1 -> 2NaCl -f H2SO4 ; (2) 2NaCl + H2SO4 -> Na2SO4 + 2HC1.
They are the two component reactions of the reversible reaction :
2NaCl + H2SO4 ^± Na2SO4 + 2HC1. Commercial saltcake (p. 238)
always contains traces of undecomposed salt and free sulphuric acid
together, because, even at a red heat, the decomposition of salt by
sulphuric acid is never complete.

Reversible and irreversible reactions. — There are a large number of
chemical reactions which appear to be irreversible under all known
conditions. Thus, magnesium burns in oxygen to form magnesium
oxide : 2Mg + O2 -> 2MgO, and even at the highest temperatures
this oxide appears to be stable. The oxidation of mercury, as in
Lavoisier's experiment (p. 47), is a similar reaction, but is reversible :
2Hg -f- O2 ^± 2HgO. Again, all organic compounds burn in oxygen
to produce carbon dioxide and water (if they contain only carbon,
hydrogen, and possibly oxygen). Thus, sugar burns in this way :
C12H22On -f- 1202-> 12CO2 + 11H20. There is no trace of sugar
left in equilibrium with C02, H2O, and O2, and the reaction is irre-
versible. Nevertheless, the reverse reaction takes_ place in-green
plantsjmdef the influence of sunlight.

These examples show that very general statements, to the effect
that all reactions are really reversible, must be accepted with reserve.
Many apparently irreversible, and to all intents complete, reactions
are, however, not so in reality. Thus, barium chloride solution is not
completely, although it is very nearly completely, precipitated by
sulphuric acid. In such cases the upper arrow will be written to
show which reaction occurs to the larger extent : BaCl2 + H2SO4 ^
BaSO4 + 2HC1. Dulong found that if barium sulphate is boiled
with successive quantities of potassium carbonate solution it is com-
pletely converted into barium carbonate ; whilst barium carbonate,
when boiled with successive quantities of potassium sulphate solu-
tion, is entirely transformed into barium sulphate : the reaction is
therefore reversible ; BaSO4 + K2CO3 ;_± BaCO3 + K2SO4. Both
BaS04 and BaCO3 are commonly supposed to be " insoluble " ;
they are, however, very slightly soluble (cf. p. 103), and the reactions

xx CHEMICAL EQUILIBRIUM, LAW OF MASS-ACTION 347

go on in solution. The solids, as such, do not in this case enter into
reaction (cf. p. 168).

The equilibrium state. — If a state of equilibrium is reached,
as a result of the balancing of two opposing reactions, it is
the same no matter which of the two groups of sub-
stances, separated by the sign ;=r, we bring together in the first
instance. Thus, the same state of equilibrium is reached on
heating hydrogen iodide to 444° for a sufficient time as is attained
on heating a mixture of hydrogen and iodine vapour, in equivalent
proportions, at the same temperature, H2 + I2 ^± 2HL This is
shown in Fig. 188. The curve A C represents the amounts of hydro-
gen iodide left after various times when that gas is heated ; the
curve EG represents the amounts of hydrogen iodide formed from
hydrogen and iodine. The former diminish, owing to the reaction :
2HI -> H2-fI2 ; the latter increase, owing to the reverse reaction :
H2 + I2 -> 2HI. Both
curves gradually coalesce to 7°
a horizontal line, CD, where
equilibrium is reached.
No further change then
occurs : H2 + I2 ^ 2HI.
Equilibrium is a state which
is independent of time. This
example shows that both
reactions can go on under
the same conditions ; in
the equilibrium state we
assume that both are still
proceeding, but the amount FIG- 188--cS£*fe.attainment °'
of hydrogen iodide formed

in any instant of time is exactly equal to the amount which is decom-
posed. The two reactions are balanced, and a state of kinetic
fUjTijlifrrinm is attai^^, "nf. ata.j-.io., when all reaction ceases.

Kinetic theory of equilibrium. — The conception of the equilibrium
state as the balance of two opposing reactions follows directly from
the kinetic theory. Thus, a liquid comes into equilibrium with its
vapour when as many molecules leap out of the liquid as return to it
in a given interval (p. 270) . A salt is in equilibrium with its saturated
solution when as many molecules break away from the solid per
second as are caught up again, possibly in a different part of the
crystal (p. 272). If barium peroxide is heated in a closed vessel
at a constant temperature, it breaks up into baryta and oxygen :
2Ba02 -» 2BaO -f- O2. The oxygen molecules, by collision with the
baryta, reproduce molecules of barium peroxide. The higher the
pressure of the oxygen, the more frequent are the collisions of oxygen
molecules on the baryta, and the greater is the rate of recombination.

•time

348 INORGANIC CHEMISTRY CHAP.

The rate at which the peroxide molecules are breaking up is constant
at a given temperature, hence at a certain pressure of oxygen the
rate at which peroxide is reproduced becomes equal to the rate at
which it is decomposed. A state of equilibrium is therefore set up
at a definite pressure of oxygen, called the dissociation pressure :
2Ba02 ^ 2BaO + O2. If the pressure of the oxygen is raised,
the collisions become more frequent, additional combination takes
place, and if the pressure is maintained above the dissociation pres-
sure, all the oxygen is reabsorbed by the baryta. If the pressure
of the oxygen is decreased, more peroxide decomposes, since less
oxygen returns to it by collisions, and if gas is continuously pumped
off, all the peroxide is ultimately decomposed. (The Brin process,
p. 168.)

Effect of volatility or insolubility of a product of reaction. — In
many cases a reaction appears to go to completion, instead of to a
state of equilibrium. Berthollet remarked that this often results from
some disturbance of the equilibrium state, by one or more of the pro-
ducts of the reaction being removed from the sphere of action by
their volatility, or insolubility. As soon as they leave the system,
passing into the gaseous state, or depositing as solids, they cease
to exert any influence, and the reaction by which they are produced,
being no longer opposed, cannot become balanced, and proceeds
until the change becomes nearly, if not quite, complete.

Thus, if sulphuric acid is poured over common salt, a state of
equilibrium is momentarily set up : NaCl -f- H2S04 ^± NaHS04 -f
HC1 f . The hydrochloric acid, however, escapes from the liquid
as a gas (shown by the upward-pointing arrow), the state of equili-
brium is disturbed, and the reaction proceeds. When the decom-
position has reached the stage where the hydrochloric acid remaining
is only sufficient to saturate the liquid, evolution of gas ceases, but
if the gas is partly expelled by heating, further reaction occurs.
Decomposition is, however, never quite complete.

If sulphuric acid is added to barium chloride solution, double
decomposition ensues : BaCl2 + H2S04 ^± 2HC1 + BaSO4 J,. The
barium sulphate, being very sparingly soluble, is precipitated (shown
by the downward-pointing arrow) ; in this way it is removed from
the sphere of action, and the reaction proceeds. The sulphate,
however, is really very slightly soluble, so that when the amount dis-
solved is in equilibrium with the solid : BaS04±^BaS04 (dissd.),
a state of equilibrium is set up. The four substances are then in
solution : BaCl2 + H2S04 ^± 2HC1 + BaS04 (dissd.) ^± BaS04
(ppd.).

Investigation of equilibrium states. — The preceding examples
show that if a state of equilibrium has been set up, it may be dis-
turbed by withdrawing one or more of the interacting substances
from the sphere of action. In examining the proportions of

CHEMICAL EQUILIBRIUM, LAW OF MASS-ACTION

840

the substances existing in equilibrium, it is also necessary to
ensure that the reverse reaction does not take place when the condi-
tions are changed. Thus, if hydrogen iodide is heated until equili-
brium is attained : 2HI ^± H2 -f I2, the proportions of HI, H2, and
I2 may be determined by rapidly cooling the mixture, when very
little reaction occurs (p. 351). In some cases, e.g., the dissociation
of steam : 2H20 ^ 2H2 + O2, this cooling must be performed
exceedingly quickly, otherwise the reverse reaction occurs, and no
trace of the products of dissociation can be discovered.

Dissociation. — The investigation of states of equilibrium attained
in the dissociation of substances by heat illustrates the point to
which reference has just been made.

Grove (1847) heated a platinum wire in steam by an electric
current. In contact with the" hot wire, dissociation occurred and
the products at once passed into the diluting atmosphere of steam,
which prevented their recombination by separating them and by

•C02+02

FIG. 189.— Deville's Experiment on Dissociation.

cooling. If a heated platinum wire (the temperature of which can
be measured from its electrical resistance) is allowed to remain for
a sufficient length of time in a flask of steam, the products of disso-
ciation and the unchanged steam are continually brought in contact
with the heated wire by diffusion, and a state of equilibrium is ulti-
mately attained, corresponding with the temperature of the wire.

Deville (1864) demonstrated the dissociation of gases at high
temperatures by means of the apparatus shown in Fig. 189. A
wide tube of glazed porcelain, with a narrower tube of unglazed
porcelain supported axially inside, was heated strongly in a furnace.
Water vapour was passed through the inner tube, and carbon dioxide
through the annular space, and the gases from both were collected
over potash solution, which absorbed the carbon dioxide. The
steam was dissociated, and the hydrogen passed out by diffusion
through the porous tube into the annular space, leaving most of the
oxygen in the inner tube. If the two gases were passed to the same
receiver, 1 c.c. of detonating gas (2H2 -f- O2) was collected for every

350 INORGANIC CHEMISTRY CHAP.

gram of water passed through the apparatus. If carbon dioxide
was passed rapidly through "a glazed porcelain tube packed with
fragments of porcelain heated in a furnace to 1200-1300°, disso-
ciation occurred : 2CO2 ^ 2CO -f- O2. When the gas was collected
over caustic potash, a mixture of carbon monoxide and oxygen was
obtained.

The effect of concentration. The law of " mass-action. "-
Berthollet, in addition to his proof of the reversibility of reactions,
made the important discovery that the extent of reaction depends
on the quantity of reacting substance present in a given volume,
or its concentration. The activity of a substance, as he says (p. 345),
is " proportional to the affinity and the quantity " ; by " quantity "
he meant " concentration." The activity is therefore proportional
to the product : (affinity) X (concentration), which Berthollet
called the active mass. A weak affinity could thus be compensated
by a large concentration, and a strong affinity weakened by high
dilution.

A substance, B, may be shared between two others, A and C,
to form AB and BC : A + EC ^ ^£^ Q If the amount of A
is increased, more of B goes to A, and a new state of equilibrium
is set up, in which the ratio AB/BC is greater than before.
Although the actual affinities of A and C for B remain unchanged,
that of A appears to have increased, because the effect of A is
proportional not only to its affinity, but also to its concentration ;
in other words, to the product of affinity and concentration, which
is called the active mass.

Thus, in reversible reactions, the extent of chemical change is propor-
tional to the active masses of the interacting substances. If to a system of
substances in equilibrium an excess of one reacting substance is
added, change occurs in such a way that the concentration of that
substance is diminished. This is known as the law of mass-action.

The law may be illustrated by an experiment due to J. H. Glad-
stone (1855). Ferric chloride and potassium (or ammonium)
thiocyanate react in solution to produce ferric thiocyanate, which
has a blood-red colour. The reaction is reversible : FeCl3 -f
3KCNS = Fe(CNS)3 + 3KC1, and if an excess of FeCl3 or KCNS
is added, the intensity of the colour deepens. But if KC1 is added,
the reverse reaction is favoured by the action of mass, and the
colour becomes paler.

EXPT. 139. — Prepare two solutions containing 2-7 gm. of crystallised
ferric chloride (FeCl3,6H2O), and 23 gm. of NH4-CNS, in 1 litre of water,
respectively. Mix 100 c.c. of each. A dark red solution of Fe(CNS)3 is
formed. Add 25 c.c. of this solution to 1 litre of water in each of four
glass cylinders ; a pale brownish-red colour is produced. Keep one
jar for reference, and to the other three add : (a) 25 c.c. of the ferric

xx CHEMICAL EQUILIBRIUM, LAW OF MASS-ACTION 351

chloride solution ; (b) 25 c.c. of the thiocyanate solution ; (c) 25 c.c. of
a saturated solution of NH4C1. Observe and explain the colour change
in each case.

Thc_concentration of a substance is usually measured by the
number of granflnolecules perJitre' Thus, if a gas mixture contains
3-62 gm. of HC1 per litre, the concentration is 0-1. Similarly, a
solution of 97 gm. of KCNS per litre has a concentration of 1. It is
convenient to denote the concentration of a substance by its
chemical symbol enclosed in square brackets, e.g., [KCNS] = 1
means 97 gm. of KCNS, or the amount represented by the formula,
in 1 litre.

Velocity of reaction. — The usual method of determining the
activity of a substance is the measurement of the rate at which a
chemical reaction involving that substance proceeds, and the
effect of change of concentration of the substance on this rate, or
velocity, of reaction.

Thus, the rate of combination of hydrogen and iodine vapour, at a
fixed temperature (e.g., 444°) : H2 -f I2->2HI, may be measured
by taking a number of bulbs containing the two substances, heating
them all to 444°, cooling successive bulbs after different intervals
of time, and determining the amount of unchanged hydrogen by
opening the bulb under potassium iodide solution. If a c.c. of
hydrogen was taken initially, and (a- x) c.c. is left after an interval
of time t, the amount of hydrogen which has taken part in the
reaction is x c.c.

In the same way, by starting with pure hydrogen iodide, and
measuring the volumes of hydrogen produced after different intervals
of time, we can find the rate of decomposition of HI.

Since we assume the velocity of reaction to.be proportional to the
activity of a substance, and the latter, at a fixed temperature, is
proportional to the concentration, it follows that the rate of combina-
tion of hydrogen and iodine vapour, i.e., the number of molecules of
H2 or I2 combining in unit time, will be proportional to the concentra-
tion of each, i.e., proportional to the product of the concentrations :

Rate of combination of H2 and I2 = kj [H2] X [I2] . . . . (1)

The constant kl is called the velocity constant : it is the rate of
combination when [HJ = [I2] = 1, i.e., when the concentrations
are unity.

In the same way, the rate of decomposition of hydrogen iodide
is proportional to the active masses. Now two molecules of HI are
decomposed : HI -f- HI -> H2 -f ^& hence :

Bate of decomposition of HI = k2 [HI]2 (2)

352 INORGANIC CHEMISTRY CHAP.

The two reactions : (a) H2 + I2 -> 2HI, (6) 2HI -» H2 -f I2,
go on simultaneously ; hence, since HI is at the same time being
formed and decomposed :

Rate of formation of HI = Rate of combination of H2 and I2 to HI
- Rate of decomposition of HI

= kt [HJ X [I2] - k2 [HI]*.

This maybe positive, negative-, or zero, according to the values
of k± [H2] x [IJ and k2 [HI]2. When the rate of formation of HI is
zero the system is in equilibrium, since then HI is decomposed
exactly as fast as it is formed, so that the amount of HI is indepen-
dent of the time. Hence in equilibrium:

*i [HJ X [IJ - k2 [HI]* = 0.
.'• ki [HJ [I2] = k2 [Hip
[HJ [IJ _ k2

" " =

At a given temperature, K is constant : it is called the equilibrium
constant. It is independent of the amounts of iodine, hydrogen, and
hydrogen iodide originally taken, but depends on the temperature.
Equation (3) is the quantitative expression of the law of mass-action
for the case under consideration. The quantitative expression of
the law is due. to Guldberg and Waage (1864).

The general equation of mass-action can now be written down.
Let the reaction :

A+B+ C + . . . ^ A' + & + C' + .'..
occur, and let it be reversible. E.g., the reaction may be :

H2 + I2 ^± HI + HI
or H2 + H2 + 02 ^ H2O + H2O,

each interacting molecule being written separately. Then the velocity
of reaction is :

^[^][B][C] . . . - i, [A'] Iff]

where [ A], etc., are the concentrations in gm. mols. per litre, and klf
are the velocity constants. In equilibrium the velocity is zero, hence :

[A] \B][C] . . . ~ i«-
where K is the equilibrium constant.

We shall always write the product of the concentrations of the
products of the reaction in the numerator, and the product of the con-
centrations of the initial substances in the denominator ; the larger the

xx CHEMICAL EQUILIBRIUM, LAW OF MASS-ACTION 353

value of K, therefore, the greater will have been the extent of the forward
reaction when equilibrium is attained.

Thus, if we consider the reaction : 3NH4-CNS + FeCl3 ^± Fe(CNS)3
+ 3NH4C1, we shall have the equilibrium equation :

[Fe(CNS),] [NH4Crp _
[FeCl3][NH4CNS]3

Addition of NH4CNS or NH4C1 will therefore displace the equilibrium
to a much greater extent than addition of the equimolecular amount
of FeCl3 or Fe(CNS)3, because the cubes of the concentrations of the
former substances are involved.

If a reaction is irreversible, there is no back -reaction, and the
speed is simply : k-± [ A] [B] [ C] . . . There cannot be equilibrium
unless one of the concentrations is equal to zero, i.e., the reaction is
complete. This agrees with what has been said of such reactions ;
they proceed to completion, without any measurable state of
equilibrium being set up.

If the reaction is reversible, the velocity constant kj_ may some-
times be measured at the beginning of the reaction, before the
products have accumulated in quantities sufficient to set up an
appreciable back-reaction. In other cases the velocity constant
can be calculated by methods which cannot be described here.
(See the author's " Higher Mathematics for Chemical Students."
Methuen.)

Kinetic derivation of the law of mass-action. — We have so far
considered the law of mass-action as an experimental fact. It may,
however, be deduced : (1) from thermodynamics ; (2) from the
kinetic molecular theory. A sketch of the second method, due to
Guldberg and Waage, will be given here.

Consider the formation of hydrogen iodide from hydrogen and
iodine. Molecules of HI can only be formed as the result of collisions
of iodine and hydrogen molecules in the gas, the number of collisions
per second being proportional to the number of molecules of each
gas present in unit volume, i.e., to their concentrations. It is
therefore proportional to the product of these concentrations,
k [H2] x [I2] • It may not be every collision which results in the
formation of hydrogen iodide, but we can assume that a definite
fraction x of the total number of collisions will be effective ; hence
the speed is equal to xk [HJ [IJ, or &a [HJ [IJ where kt= xk, and
x, k, are constants. Similarly, the speed of the reverse reaction will
be k2 [HI]2, since two HI molecules must collide, and the probability
for this is proportional to [HI]2.

Those molecules which are in a condition to undergo chemical change
on collision (active molecules) appear to be those possessing more
than a certain critical amount of internal energy, due to atomic rotation

A A

354 INORGANIC CHEMISTRY CHAP.

or vibration. When a molecule acquires this critical increment of
internal energy it becomes active

Thermal dissociation. — Let 2 gm. mol. (254 gm.) of hydrogen
iodide gas, contained in a sealed bulb of volume V litres, be heated
at 444° in the vapour of boiling sulphur. After a certain lapse of
time a state oi equilibrium is attained :

HI + HI ^± H2 + I2.

Let the fraction y of the hydrogen iodide be dissociated ; the
equilibrium concentrations will then be as follows :

HI + HI ^ H2 + I2
[HI] [HI] [HJ [IJ

lny l—y ^L 2.

v V v v

The law of mass-action states that the product of the concentra-
tions of the substances formed in the reaction, divided by the pro-
duct of the concentrations of the original substances which remain,
is equal to a constant, K, at a given temperature :

y_ y_

ft] XJHJ - V ' V _ y*
[Hlfx [HI] 1 -^ 1-- y (1 - y)2 -

V V

The resulting equation does not contain F, so that the degree of
dissociation y, of HI, is independent of the volume of the bulb in
which the HI was initially confined ; in other words, it is independent
of the pressure, and depends only on the temperature.

EXAMPLE. — 7 94 c.c, of hydrogen (at S-T.P.) and 0-0601 gm. of solid
iodine were heated in a sealed bulb at 444° until equilibrium was reached.
9-52 c.c. of hydrogen iodide were formed. Now at S.T.P. 2 x 126
gm. of iodine (I2) occupy 22,240 c.c.

.*. vol. of I2 vapour at S.T.P. initially present
22240 X 0-0601

—* X 126 =5'30c'c-

Each c.c. of HI formed diminishes the volume of the H2 and I2 by
0-5 c.c. each, .*. in equilibrium :

vol. of H2 = 7-94 - 4-76 = 3-18 c.c. (4-76 = 0-5 X vol. of HI =

0-5 X 9-52)

vol. of I2 = 5-30 — 4-76 = 0»54 c.c.

vol. of HI = 9*52. Hence, if V is the volume of the bulb, the
concentrations are: [H2] = 3-18/22240F; [I2] = 0-54/22240F;
[HI] =4.76/22240F.

K - ^ X M _ 3- 18 X 0-54 _
~ ~ ~ (4-76)2

xx CHEMICAL EQUILIBRIUM, LAW OF MASS-ACTION 355

Now, suppose 8-10 c.c. of hydrogen and 2-94 c.c. of iodine vapour (at
S.T.P.) heated to 444°. What volume of HI will be formed in equili-
brium ? Let 2x c.c. be formed :

H2 + I2 — HI + HI

Volumes : (8-10 - x) (2-94 - x) x x

.-. x = 2-822 or 9-12. Only the root 2-82 is admissible, since 2-94 c.c.
I2 vapour can give only 5-88 c.c. HI as a maximum. Thus, volume
of HI formed = 2 X 2-82 c.c. = 5*64 c.c. Bodenstein by experiment
found 5-66 c.c.

Effect of temperature and pressure on equilibrium. — The

dissociation of hydrogen iodide cannot be measured by the
change of density, because the volume is unchanged. But if an
increase of volume occurs, the extent of dissociation can be measured,
as described on p. 152, from the vapour density. This is the case
with phosphorus pentachloride:

PC15 = PC13 + C12

1 — y y y

Concentrations : — y — -y- -£-

= K.

The extent of dissociation now depends on the volume, V, and
therefore on the pressure, which was not the case with HI, since in
the latter case V cancelled out in the equilibrium equation.

If V is increased (i.e., the pressure diminished), the denominator in
the above expression for K becomes too large ; the numerator,
and therefore y, must also increase in order to maintain the value
of the equilibrium constant. Hence the extent of dissociation
increases, in this reaction, when the pressure is reduced. The same
effect is produced by adding an indifferent gas, which reduces the
partial pressures. A change of volume or pressure influences the
state of equilibrium only when the chemical reaction causes a
change of volume (e.g., PC15 (1 vol.) = PC13 + C12 (2 vols.)). If no
change of volume occurs (e.g., 2HI = H2 -f- I2), pressure has no
influence on the equilibrium. If the pressure on a system in equili-
brium is increased, that change occurs which leads to a diminution
)f volume, i.e., a decrease of pressure, and the equilibrium is corre-
spondingly shifted. This is a special case of the law of reaction : if

system in equilibrium is subjected to a constraint, a change occurs,
if possible, of such a kind that the constraint is partially annulled,
The effect of pressure on equilibrium is so regulated.

Another aspect of this law is the effect of temperature on equili-

A A 2

356 INORGANIC CHEMISTRY CHAP.

brium. If the temperature of the system in equilibrium is raised
(or lowered), that one of the two reversible reactions will occur
which absorbs (or evolves) heat. Thus, the dissociation of PC15
is increased by raising the temperature, because the reaction
PC15 = PC13 + C12 occurs with absorption of heat.

If Qv is the heat of reaction at constant volume (p. 387), and Klt K% are
the equilibrium constants corresponding with the absolute temperatures
Tl and T2, then it is shown by thermodynamics that, if 1 gm. molecule
of substance is taken :

In this way the heat of reaction may be calculated.

EXAMPLE. — 2-0 gm. of PC15 are sealed in an evacuated bulb of 200 c.c.
capacity, heated to 200°. Find, the pressure developed if PC15 is 48-5
per cent, dissociated under 1 atm. pressure.

2-0 gm. of PC15 = 2-0/207 = 0-0097 gm. mol. Let x = degree of
dissociation under conditions of experiment. Let the volumes be
measured in litres ; then

°-°097- x x *2

- - 6-20 ' s " a 6-20 •• (

But at 200° under 1 atm. pressure PC15 is 48-5 per cent, dissociated.
The volume of 1 gm. molecule under these conditions is

22-24 x 1-485 X = 57-2 lit.

'• K - Onn* - (00097-*) .

/. z=0f0033. There are thus 1-0033 X 0-0097 gm. mol in 200 c.c.

.'. pressure = 1-00973 x ~| x ^^ = 1'93 atm.

(If the pressure is doubled, the dissociation diminishes from 48 '5 to
0-33 per cent.)

Effect of addition of products of dissociation. — Let 2 gm. mol. of
hydrogen iodide, contained in a volume V, be dissociated to the
extent y. Now suppose x gm. mol. of hydrogen (or iodine vapour)
introduced into the vessel, at the same temperature. The original
concentrations (see p. 354) were : [HI] = (1 — y)/F; [HJ = y/F ;
[IJ — y/F, andy2/(l — y)2 = K. When the excess of hydrogen is,
added, [HJ = (y + x)/V. The product is now y (y + a;) /(I — y)2,
which is greater than K. Hence y must diminish to a
value y', such that y (y + x)/(l — y')2 = K. The extent
of dissociation is therefore diminished by adding hydrogen or iodine
vapour at constant volume.

Now suppose a volume nV of hydrogen (or iodine vapour) added,

xx CHEMICAL EQUILIBRIUM, LAW OF MASS-ACTION 357

at the same concentration as it exists in the mixture, i.e., V contains
y gm. mol. The concentrations are now :

[HI] = (1 - y)/(l + n)V ; [H2] = y(l + n)/V(l + n) = y/V ; [IJ

The product is (1 -}- n) y2/(l — y)F, which is greater than K.
The value of y must therefore diminish, i.e., the dissociation is
diminished, although the volume is increased.

In the case of the dissociation of PC15, the addition of x gm. mol.
of PC13 (or C12) at constant volume changes the concentrations in the
mixture to : [PC1J = (1 - y) /V ; [PC1J = (y -f x) /V ; [01 J =
y/V. Their product is: (y -f x)y/(l — y)F, which is larger
than the equilibrium value. The extent of dissociation, y, must
therefore diminish on adding one of the products of dissociation
-when the volume is constant.

But if nV volumes of PC13 (or C12) are added at the same concentra-
tion as it exists in the mixture, i.e., V volumes contain y gm. mol.,
then the concentrations are :

[PC15] = (1 - y)/F(l + n) ; [PC1J = y(l + n)/F(l + n) ; [C1J =
y/F(l + n),

and their product is y2/(l — y)F, i.e., K, so that no change is
produced.

Since there are now 1 + y (1 + n) molecules in a volume F^l -f- n),
whereas before the addition of PC13 or C12 there were 1 -f y molecules

a volume F, i.e., (1 -f y)(l + n), or 1 -f- y(l -j- n) + n, molecules

a volume F(l + n), it follows that the total pressure is reduced
by addition of a product of the dissociation at the same partial
pressure as it exists in the mixture. If this substance is added so
that the total pressure remains constant, it follows that the dissocia-
tion will be reduced. This was shown by Wurtz (1873) : if PC15
is volatilised into an atmosphere of PC13 at atmospheric pressure,
the dissociation is largely suppressed, and only a very pale greenish
colour, due to chlorine, is seen.

Electrolytic dissociation. — If a salt, acid, or base is dissolved in
water, its molecules are partially broken up into ions, the degree
of ionisation, a, increasing with dilution to a limiting value 1, when
dissociation is complete : NaCl ^ Na* -f 01'. Let 1 gm. mol. of electro-
lyte be dissolved in a volume F. The concentrations are then :

[NaCl].= (1 - a)/F ; [Na*] = [01'] = a/F.
If the law of mass-action applies to ionisation we have :

[Nad] ~ (1 - «)F =
This equation is known as Osj&aldXjlilution law (1886)^ It
applies with very great exactness to weafc~electroiytes, as may be

368 INORGANIC CHEMISTRY CHAP.

seen from the values of K, the ionisation constant, for acetic acid
on p. 291. In the case of strong electrolytes, however, it fails com-
pletely, as may be seen from the values of K for potassium chloride
on p. 291. The reason for the anomalous behaviour of strong elec-
trolytes is not known.

Solubility product. — If solid sodium chloride is in contact with its
saturated solution we have two connected equilibria :

NaCl (solid) z± NaCl (dissd.) =± Na* + Cl'.

The total solubility of sodium chloride is the sum of the amounts of
the undissociated NaCl molecules (sometimes called the true solu-
bility), and of the NaCl dissociated into ions, Na* and Cl'.

If the law of mass-action applies to the ions (cf. above) we have :
[Na'] x [Cl'] - JC[NaCl]. But [NaCl] is the concentration of
un-ionised salt, i.e., the true solubility. It is assumed that this is"
always constant at a given temperature, if excess of solid is present.
Hence, in equilibrium, the product of the ionic concentrations is
constant at a given temperature. This constant product, e.g.,
[Na'] x [CF], is called the solubility product. When the product
of the ionic concentrations, or the ionic product, is equal to the solu-
bility product, the solution is in equilibrium with the solid, since
then the concentration of un-ionised salt in solution must be that
which is in equilibrium with solid. If the ionic product is less
than the solubility product, the solution is unsaturated with re-
spect to the solid, and more of the latter dissolves. But if the
ionic product is greater than the solubility product, the solution is
supersaturated, and precipitation of solid usually occurs. In some
cases the solution remains supersaturated.

The value of the ionic product may be increased by adding to the
solution an electrolyte which has an ion in common with the sub-
stance in solution. Thus, if hydrochloric acid is added to a saturated
solution of common salt, the concentration of chloride ions is
increased, and the ionic product, [Na'] x [Cl' from NaCl -f- added
Cl' from HC1], is increased above the value corresponding with the
solubility product. Solid sodium chloride is then precipitated until
the ionic product becomes equal to the solubility product, i.e.,
until the concentration of the un-ionised salt in solution is reduced to
its original value, the true solubility. The other ion of the added
electrolyte, H*, has no effect on the equilibrium, as may be proved
by adding a quantity of another chloride, e.g., LiCl, containing the
same quantity of chloride ions as the acid, when the same weight
of NaCl is precipitated as in the first experiment. If an equivalent
amount of Na* ions, e.g., as NaC103, had been added instead of Cl'
ions, the effect is exactly the same, as it should be, since the product
[Na'] x [CF] is affected to the same extent by equivalent amounts
of Na* and Cl'.

xx CHEMICAL EQUILIBRIUM, LAW OF MASS-ACTION 359

EXPT. 140. — Pass gaseous hydrochloric acid into a filtered saturated
solution of common salt, which contains magnesium chloride as im-
purity, using the apparatus shown in Fig. 190 A white crystalline
powder of NaCl falls. This is filtered off, washed with a little pure
concentrated hydrochloric acid, and, after removing mother liquor in
a Biichner funnel, is
dried on a porous plate.
The salt is heated care-
fully in a platinum dish
to drive off HC1, and
is then pure.

EXPT. 141. — To a
saturated solution of
silver acetate add :
(a) a concentrated solu-
tion of silver nitrate :
(6) a saturated solu-
tion of sodium acetate
In each case silver
acetate is precipitated.

EXPT. 142.— To a
saturated solution of
potassium perchlorate
add : (a) perchloric
acid ; (6) a saturated
solution of potassium
chloride ; (c) a satur-
ated solution of sodium
chloride ; (d) concen-
trated hydrochloric
acid. Explain the effect
in each case. In
case (d), if HC1 is
" hydrated " in solu-
tion, it would be ex-
pected to withdraw "free water " from the solution of the salt. From
the result, consider whether this " explanation " covers EXPT. 140.

The effect of adding a slight excess of a reagent in analytical
chemistry is now clear. If exactly equivalent amounts of silver
nitrate and hydrochloric acid in aqueous solution are mixed, precipi-
tation of silver chloride occurs : Ag' -f Cl' ^± AgCl \|/. If the
precipitate is filtered off, and either silver nitrate or hydrochloric
acid added to the clear filtrate, an opalescence is produced, owing
to precipitation of AgCl. A trace of the latter existed in solution,

FIG. 190.— Preparation of Pure Sodium Chloride.

360 INORGANIC CHEMISTRY CHAP.

almost completely ionised at the great dilution, and when a common
ion was added the ionic product, [Ag'] x [Cl'], exceeded the solu-
bility product. In the quantitative precipitation of silver, or of
chlorides, a slight excess of a chloride, or of silver nitrate, respec-
tively, is added. The precipitation is then practically complete.

If a large excess of concentrated hydrochloric acid is added to the
precipitate of silver chloride, some of the latter dissolves. In this case,
however, the substance in solution is not AgCl but a complex substance,
probably H2AgCl3, which ionises as follows : H2AgCl3 ^± 2H + AgCl3".
The solution contains no silver ions, Ag*, and the phenomenon is not
an exception to the solubility product equation. As a general rule,
too great an excess of reagent should not be added in precipitation
reactions.

EXPT. 143. — To a solution of silver nitrate add drop by drop a solution
of potassium cyanide, KCN. A white precipitate of silver cyanide is
first produced : Ag' -f ON' ^± AgCN |. On continued addition of the
cyanide this precipitate redissolves, and the solution then contains the
complex anion Ag(CN)2' : AgCN + CN' ^± Ag(CN)2'. There are
then present 2K", NO3', and Ag(CN)2', i.e., KNO3 and KAg(CN)2.
In the latter salt, potassium argentocyanide, the silver is present in the
acid radical, or anion, and the solution is practically free from silver ions,
Ag". It gives, for instance, no precipitate of AgCl with a soluble
chloride. Complex ions are slightly broken up into the simple ions in
solution. Thus, the reaction Ag(CN)2' rz; Ag" + 2CN' occurs to a slight
extent, and silver is deposited on the cathode from this solution in
electroplating (p. 825).

Hydrolysis.— A number of salts are decomposed to a greater or
less extent on solution in water, with the separation of acid and base.
The reaction is known as hydrolysis.

Thus, a solution of potassium cyanide has an alkaline reaction,
and smells of hydrocyanic acid, owing to hydrolysis : KCN -f
H2O -^ KOH + HCN.

EXPT. 144. — Heat about 5 gm. of plaster of Paris with half its weight
of powdered charcoal in a covered crucible. Calcium sulphide is
formed : CaSO4 + 4C = CaS + 4CO. When cold, add the solid to
water. The liquid will be found alkaline to litmus, and, on warming,
sulphuretted hydrogen is evolved, turning lead acetate paper black :
CaS + 2H2O ;=± Ca(OH)2 + H2S. . H. Rose (1842) found that the hydro-
lysis in this case increases with the dilution.

EXPT. 145. — To a concentrated solution of borax add silver nitrate. A
white precipitate of silver metaborate is formed :

Na2B4O7 + 3H2O z± 2NaBO2 + 2H3BO3 ;
NaBO2 + AgNO3 = AgBO2 + NaNO3.

xx CHEMICAL EQUILIBRIUM, LAW OF MASS-ACTION 361

Repeat the experiment with a very dilute solution of borax : a brown
precipitate of silver oxide is obtained :

NaB02 4- 2H20 ^ NaOH + H3BO3 •

2NaOH + 2AgNO3 = 2NaNO3 + Ag2O + H2O.

In the above examples, the salts are formed from a weak acid and
a strong base. E.g., H2S and Ca(OH)2 ; H3BO3 and NaOH ; HCN
and NaOH (cf. p. 422). In solution, the weak acid is scarcely ionised
at all, the slight ionisation which would occur in pure water being
almost completely suppressed by the action of the anion, which is
produced from the largely ionised salt : HCN ^ H' -|- CN' ;
KCN ;=± K' -f CN'. The strong base, on the other hand, is largely
ionised : KOH ^± K' -j- OH' ; so that the solution exhibits : (i) the
properties of the free acid, (ii) a strongly alkaline reaction, due
to hydroxide ions from the strong base.

With salts of a strong acid and a weak base, the results are the
opposite of those just described.

EXPT. 146. — Test with litmus paper solutions of copper sulphate and
ferric chloride : notice the acid reaction.

EXPT. 147. — Pour a dilute solution of ferric chloride on a parchment
paper dialyser (p. 314), and float on water. The water becomes acid,
hydrochloric acid diffusing through the membrane, and colloidal ferric
hydroxide is left in the dialyser : FeCl3 -f 3H2O ^± Fe(OH)3 + 3HC1.

In a solution of ferric chloride, the ferric hydroxide, Fe(OH)3,
which exists in the state of a colloidal solution (cf. p. 989), is a very
weak base, practically not ionised, whilst the hydrochloric acid is
largely ionised. Hence the solution has an acid reaction, from the
presence of hydrogen ions : HC1 z=± H" -f Cl' . The dark brown
colour of the dialysed solution is due to the un-ionised base
Te(OH),.

In a solution of a salt of a weak acid with a weak base hydrolysis
also occurs. Thus, a solution of ammonium carbonate is alkaline,
because, although ammonia is a weak base, it is stronger than car-
bonic acid : (NH4)2CO3 + 2H2O ^± 2NH4OH -f H2C03 ; NH4OH
=± NH4' + OH' ; H2C03 ^ HC03' + H .

Very slight hydrolysis probably occurs in solutions of all salts :
it may become appreciable at high temperatures.

EXPT. 148. — Heat a little NaCl in a platinum crucible to redness, and
add a few drops of water. These assume the spheroidal state. After a
few seconds transfer the drop of water to a beaker containing distilled
water faintly coloured with litmus : this is turned red. Allow the
crucible to cool, dissolve the salt in water, and add to dilute red litmus
solution : this is turned blue. NaCl + H2O ^ HC1 + NaOH.

362 INORGANIC CHEMISTRY CHAP.

The law of mass-action may be applied to hydrolytic reactions :

[acid] X [base] _
[salt] X [water] ~

Hydrolysis may be considered as due to the action of the ions
of water. If sodium hypochlorite, for instance, is dissolved in water,
it first of all ionises : this occurs to a considerable extent, since
nearly all salts are largely ionised in solution (p. 294) : NaOCl ^
Na'-|- OCr. The ion OC1' thus finds itself in the presence of a very
small concentration of hydrogen ions derived from the ionisation of
water. Combination then occurs between the OC1' ions and H'
ions to form undissociated HOC1, since the ionisation of this acid is
so slight, especially in presence of the large number of OC1' ions, that
the concentration of H' ions derived from it : HOC1 ±ir H* -f- OC1',
is less than the concentration of H' ions derived from water :
H2O ±£ H' + OH'. By reason of the removal of H' ions,
further ionisation of water occurs, and the reaction H' -(- OCl'^^
HOC1 proceeds until the concentration of HOC1 in the solution is
such that the H' ions formed by its excessively slight dissociation
are in equilibrium with the OH' ions produced from the water :
[IT] x [OH']— [H2O] = const. It is these OH' ions which
give the solution its alkaline reaction. The reactions may be
summarised as follows :

NaOCl ;=±Na' + OC1'

H20±^H' + OH'

•H' + OC1' ^± HOC1 nearly undissociated ;

which give, on addition :

NaOCl + H20 =± Na' + OH' + HOC1.

Now suppose that both the acid and base are weak. The ions of
the salt now react with both the ions of water to form nearly un-
dissociated acid and base. Thus, with ammonium hydrosulphide,
we obtain ammonia and sulphuretted hydrogen (a weak acid) :

NH4-HS^±NH4* + HS'

H20 ^ H- + OH'
H + HS' ^± H2S

which give, on addition :

NH4-HS + H20 — NH4-OH + H2S ; or
NH4- + HS' + H20 = NH4-OH + H2S

since NH4'HS is largely ionised in solution.

The reaction of the solution will now depend on the relative
strengths of the acid and base. Since combination of both the
anions of the acid, and the cations of the base, with H' and OH' ions

XX CHEMICAL EQUILIBRIUM, LAW OF MASS-ACTION 363

of water, respectively, occurs, the hydrolysis is greater in this case
than when only the acid, or base, alone is weak.

Theory of indicators. — The action of acids and alkalies in changing
the colour of certain substances has long been known, and utilised
in testing for these two groups of compounds. Many natural
colouring matters may be used for this purpose, the most important
being litmus, a colour prepared from certain lichens (Eoccella
tinctoria, Lecanora Tartar ea, etc.) growing on rocks near the sea.

Litmus comes into the market in small cubes, of a blue colour. These
are powdered, and digested on a water -bath with water to which about
one-fourth its volume of alcohol is added ; the deep blue solution is
decanted from the residue (calcium carbonate), filtered, and the free
lime in the solution neutralised with dilute sulphuric acid until, after
boiling, the solution has a purple colour. Filter paper soaked in the
solution and dried forms litmus paper.

Turmeric papers (from an alcoholic extract of the ground root of
the Curcuma longa, of India, used in making curries) are yellow,
turned reddish-brown by alkalies or boric acid (p. 738). A
number of synthetic organic substances are now also used as
indicators.

Methyl-orange in aqueous solution is turned yellow by alkalies
and red by acids ; paranitraniline is colourless in acid solution,
yellow in alkaline solution ; methyl-red is turned red by traces of
acids, and yellow by alkalies ; phenolphthalein is colourless in acid
solution, and is turned deep red by traces of alkali ; alizarin red
is turned deep purple by alkalies, yellow by acids.

According to Ostwald's theory of indicators (1894) these substances
are weak acids or bases, one radical of which, in the ionic state, has
a different colour from that in the undissociated molecule. Thus,
paranitraniline is a weak acid, which in the undissociated state
is colourless. A trace of strong acid added to its solution drives
back the slight dissociation of the weakly acidic indicator in the
aqueous solution, and the pale yellow solution becomes colourless.
If, however, an alkali is added, the OH' ions combine with the H"
ions of the indicator to form H2O molecules, and further ionisation
of the weakly acidic indicator occurs. The anion of the indicator
then exists in the solution in appreciable amounts, exhibiting a
strong yellow colour.

In many of these reactions changes of structure, i.e., of valency and the
arrangement of the atoms, may occur in the radical when it leaves the
neutral molecule to form an ion ; this does not necessarily affect the
above theory of indicators.

Phenolphthalein is supposed to function as a very weak acid ; its
salts, formed by the action of alkalies, are largely dissociated, giving

36* INORGANIC CHEMISTRY CHAP.

an intensely red anion. Its action is similar to that of ^p-nitro-
aniline. Methyl-orange functions as a very weak base ; its solutions
contain traces of OH' ions and a red cation, whilst the undissociated
substance is yellow, so that the aqueous solution of the indicator
is orange-red. On addition of alkali, the ionisation of the
indicator is forced back, and the yellow undissociated mole-
cules are formed. The H' ions of acids combine with the OH'
ions of the indicator to form H20 ; further ionisation of the
indicator takes place, and the red colour of the cation appears :
X-OH±^X' (red) + OH'.

Sensitiveness of indicators. — An indicator requires a definite
concentration of H* or OH' ions to produce its characteristic colour
change : this concentration varies with different indicators. Thus,
methyl-violet is turned blue by a definite small concentration of
strong acids (e.g., H2S04), whereas it is unchanged by the weak
acetic acid at any concentration, since the latter can never produce
the requisite concentration of H' ions.

The ionic product [H*] x [OH'] is constant in all aqueous
f

solutions on account of the equilibrium: H2O^H' + OH', and
equal to the dissociation constant of water : [H'] x [OH'] =
10~138. The concentration of OH' ions required to produce
a colour change of an indicator may, therefore, always
be represented by the equivalent H' ion concentration :
[OH/] = [OH']x[H-]/[H*] = 10-138/[H<]. At the neutral point the
H' and OH' concentrations are equal, each being equal to its con-

centration in' pure water : [H'] = [OH'] = VlC'13'8 = lO"69. If
[H'] is greater than 10~69 the solution is acid ; if it is less than
10~6* , e.g., 10~8, the solution is alkaline. The concentration of
H' ions may be represented by minus the exponent of the con-
centration, and is then usuajly written pH. ; e.g., if [H'] = 10~81,
pH.=S'l. This is called the sensitiveness of the indicator. An
ideal indicator, which shows the exact point of neutrality, corre-
sponds with £>H.=6-9.

The values of the H* ion concentrations required to pro-
duce colour changes of various indicators are given in the table
below, compiled from the results of Salm (1906) and Sorensen
(1909).

The gaps are to be filled in with the colour next adjoining, e.g.,
phenolphthalein is colourless with all H' concentrations greater than
10~8, red for all less than this. It will be seen that litmus approaches
an ideal indicator, i.e., a solution reacting neutral to litmus is actually
neutral : [H'] = [OH'] = I0~6'9 ; with phenolphthalein the solution
would be still faintly alkaline ; with methyl -orange it is slightly acid
(0-0005AT) ; whilst methyl -violet requires 0-01A7-acid to produce a
colour change.

CHEMICAL EQUILIBRIUM, LAW OF MASS-ACTION

365

i !

Colour of Indicator with Hydrogen-ion concentration normal multiplied by : —

Indicator.

Methyl-
violet.

Methyl-
orange.

Congo
red.

Methyl-
red.

Litmus.

Phenol-
phthalein.

Golden -
yellow

Green

10-1

Greenish-
blue

10-2

Blue

Red

10-.

Violet

Orange-
red

Blue

Violet-
red

10-4

Orange-
yellow

Violet

Red

10-5

Scarlet

Orange-
yellow

10-6

Yellow

Red

10-7

Blue

Colourless

10-8

Rose-red

10-9

Red

10-io

p.c. Indicator
solution.

0-05

0-01

0-2 p.c. in
60 p.c.
alcohol

—

0-05 in 50
p.c.
alcohol

Drops indicator to
10 c.c. test.

3-8

5-10

3-5

—

3-20

EXPT. 149. — Three rows of five small flasks, each containing 100
c.c. of " conductivity " water (p. 212), are supported on a rack (Fig. 191)
with milk-glass shelves. To the flasks of each row are added :
p-nitrophenol, methyl-orange, litmus, phenolphthalein, and methyl-red,
respectively. To the top row (A) a drop of baryta water is added,
when the indicators give the alkaline reaction. To the bottom row ( C)
a drop of jVH2SO4 is added; when the indicators give the acid reaction.
To the middle row (B) 1 c.c. of very dilute baryta water is added from a
burette, and then, by means of a series of small tubes fastened to a board,
as shown, 1 c.c. of freshly distilled methyl formate is poured simulta-
neously into all the flasks of this row. The methyl formate slowly hydro-

366

INORGANIC CHEMISTRY

lyses, giving methyl alcohol (neutral) and formic acid : H-COO-CH3 -f
H2O = H-COOH (formic acid) + CH3'OH (methyl alcohol). The H'
ions of the formic acid neutralise the OH' ions .of the baryta, and then
excess of H' ions are formed. The solutions therefore change over from
alkaline, through the point of exact neutrality, to acid. If the point

of neutrality is taken as that
corresponding with the colour
change of litmus, the reactions
of the other indicators, which
change at different times, may
be compared (Nernst, 1908).

ff &

FIG. 191. — Experiment on Indicators.

SUMMARY OF CHAPTER XX

The cause of chemical change
is identified with the affinities
of the interacting substances,
which may be electrical forces.
The activity of a substance,
which may be measured by the
velocity of reaction, was shown
by Berthollet (1799) to depend
not only on its affinity, but also
on its concentration, i.e., the number of molecules in unit volume. The
product of affinity and concentration is called the active mass.

The Law of Mass- Action states that the activity is proportional to the
active mass, i.e., to the concentration. The product of the concentrations
of the substances produced, divided by the product of the concentrations of
the interacting substances, when equilibrium is attained, is constant :

A + B + C + . ..^D+E + F...

[A] X [B] x[C]
K is called the equilibrium constant.

EXERCISES ON CHAPTER XX

1. Point out the fallacy in the statement : " Potassium chlorate
contains chlorine, although it gives no precipitate with silver nitrate."

2. Describe briefly the history of the theory of affinity until the time
of Berthollet. What modifications did the latter introduce into the
theory ? It has been stated that : " the phrase ' active mass ' com-
monly employed instead of the words ' molecular concentration ' . . .
is distinctly misleading." Criticise this assertion.

3. What experiments would you carry out in order to determine if
the precipitation of bismuth chloride by water : BiCl3 + H2O = BiOCl
+ 2HC1, (a) is reversible ; (b) is subject to the law of mass-action ?
The law of mass-action strictly applies only to homogeneous systems :
defend its use in the present instance (cf. p. 358).

xx CHEMICAL EQUILIBRIUM, LAW OF MASS-ACTION 367

4. What would be the effect of increasing the pressure on the following
systems in equilibrium : N3O4 ^± 2NO2 ; H2O + CO ^± H2 + CO, ;
2S02 + 02 =± 2S03 ; N2 + O2 ^ 2NO ; N2 + 3H2 =± 2NH3 ? What
the general law relating to such cases ?

5. If electric sparks are passed through a mixture of nitrogen and hy-
drogen, a little ammonia is formed : N2 -f 3H2 ^ 2NH3. How would
you arrange an experiment in which a mixture of nitrogen and hydrogen
is to be completely converted into ammonia ?

6. On what experimental evidence is it believed that acids, bases, and
salts are ionised in aqueous solution ? Explain from this point of view
(a) the alkaline reaction of sodium carbonate solution, (6) the acid
reaction of copper nitrate solution.

7. It was considered that the atomic theory was fundamentally
opposed to Berthollet's theory that mass produces an effect on chemical
reactions (cf. p. 111). Explain how the two may be reconciled.

8. State briefly the modern theory of acids. Why are HC1 and
H2SO4 acids, whilst NH3 and NaOH are not ? Discuss the place
occupied by water, H2O.

9. A solution of silver nitrate is added drop by drop to a solution of
hydrochloric acid. In what way does the very slight solubility of
silver chloride vary as the reaction proceeds up to and beyond the point
when an equivalent amount of silver nitrate has been added ?

10. Discuss the theoretical foundation for the statement that endo-
thermic compounds are more stable at high temperatures. Give
examples. It is sometimes assumed that, at the temperature of the
sun (6000° — 10,000°) all compounds must be dissociated into their
elements. Criticise this.

11. Explain how, when an acid solution is titrated with an
alkaline solution, the neutral point may be determined from measure-
ments of the electrical conductivity, without an indicator. Does the
end-point obtained with an indicator necessarily indicate exact neutral-
ity ? How is the latter defined ?

12. What explanation can you give of the fact that, although
phenolphthalein is very sensitive to bases, it is insensitive to ammonia
(a weak base) ? Why may methyl -orange be used ?

13. What is meant by strong, and weak, acids ? Arrange the
following acids in the order of strength : acetic, nitric, sulphuric,
carbonic, hydrochloric, hydrocyanic. What effects would aqueous
solutions of the sodium salts of these acids have upon red and blue litmus
papers ?

CHAPTER XXI

THE OXIDES AND OXY-ACIDS OF CHLORINE

The action of chlorine on alkalies. Hypochlorites. — If a
stream of chlorine is passed through a cold dilute solution of
caustic potash, a liquid smelling somewhat like chlorine, but
with a distinct difference, is obtained. This liquid, discovered
by Berthollet in 1789, possesses bleaching properties, and since
it is more stable than chlorine water, it was used, under the
name of eau de Javelle, for bleaching. This solution began to
be used in England about 1798, but the absorption was carried out
with milk of lime instead of with potash. Tennant, of St. Rollox
(Glasgow), in 1799 found that chlorine is absorbed by dry slaked-
lime, and the product, called bleaching powder, on treatment with
water, gave a bleaching liquor.

The composition of these bleaching substances was investigated
in 1842 by Balard, who showed that they contain salts of hypo-
chlorous acid, HOC1. The reactions mentioned lead to the forma-
tion of a mixture of a hypochlorite and a chloride :

2KOH + C12 = KC1 + KOC1 + H20.
2Ca(OH)2 + 2C12 = CaCl2 -f Ca(OCl)2 + 2H2O.

Instead of caustic potash, the cheaper caustic soda may be used,
when a solution containing sodium hypochlorite, NaOCl, and sodium
chloride, is formed. This solution is also produced by adding
sodium carbonate solution to a solution of bleaching powder, and
filtering off the precipitated calcium carbonate :

Ca(OCl)2 + CaCl2 + 2Na2C03 = 2NaOCl + 2NaCl + 2CaC03,

or, more usually, by the electrolysis of brine under special con-
ditions, so that the chlorine liberated at the anode is allowed to
mix with the caustic soda produced at the cathode, and the liquid
is kept cool.

The bleaching action of hypochlorites is due to free hypochlorous
acid, HOC1, liberated by acids :

NaOCl + H2S04 = NaHS04 + HOC1.

CH. xxi THE OXIDES AND OXY-ACIDS OF CHLORINE 369

Even carbonic acid, e.g., atmospheric carbon dioxide, turns out the
very weak hypochlorous acid from its salts ; hence solutions of these
smell of the- free acid when they have been exposed to air, and
exhibit bleaching properties.

EXPT. 150.— Pass chlorine into cold dilute caustic soda solution.
Take a piece of Turkey red cloth and paint on it ,a device with a mixture
of gum and tartaric acid. Dry the cloth in a steam-oven and then
immerse in the hypochlorite solution (containing a slight excess of alkali).
The colour is discharged only where the acid was applied. Now pass a
stream of carbon dioxide through the liquid : the colour is now com-
pletely discharged : NaOCl + CO2 + H2O = NaHCO3 -f HOC1.

The bleaching action of hypochlorous acid is due to oxidation :
HOC1 = HC1 -f 0. Many colouring matters when oxidised yield
colourless or feebly-coloured products.

Thus, indigo blue, C16H10N2O2, yields the yellow isatin, C8H5NO2 .
C16H10N202 + 2HOC1 = 2C8H5NO2 + 2HC1. The yellow colour of
unbleached cotton or linen is due to a natural brown colouring matter.
In bleaching the yarn or fabric, it is first boiled with dilute caustic soda,
to remove oily and resinous substances, and some colour. It is then
washed, immersed in bleaching powder solution, and finally in dilute
sulphuric acid, or exposed in piles to the air (carbonic acid). The acid
sets free hypochlorous acid. The cellulose, of which the cotton fibres
are composed, is resistant, unless the action is too prolonged, but the
colour is oxidised. The remaining hypochlorous acid is removed by
washing, and finally by treating with a substance such as sulphur
dioxide, which decomposes the hypochlorous acid, and is hence called
an antichlor (p. 522) : HOC1 + SO2 + H2O = HC1 + H2SO4. Paper
pulp, prepared from wood, is bleached with sodium hypochlorite solu-
tion and acid (p. 847).

Chlorine water. — The bleaching action of chlorine water may also
be regarded as due to the hypochlorous acid it contains, although
a considerable amount of free chlorine is present, since the reaction :
C12 + H20 ^ HOC1 -f HC1 is reversible.

The following equations :

C12 + H2O = 2HC1 + O
HOC1 = HC1 + O

show that hypochlorous acid, for the same weight of chlorine, has
twice the bleaching activity of free chlorine. There is therefore no
loss of bleaching activity when the chlorine is first absorbed by
alkali, although half the chlorine is converted into inert chloride.
It is the available oxygen liberated from HOC1 which causes the
bleaching action.

The constitution of chlorine water, explained above, may be

B B

370

INORGANIC CHEMISTRY

CHAP.

proved by the following experiments. If chlorine water is distilled,
hypochlorous acid co'mes over, leaving aqueous hydrochloric acid.
In this case the equilibrium : C12 + H20 ^ HOC1 -f HC1, is dis-
turbed by the removal of the volatile constituent HOC! (or its
anhydride, C120 : 2HOC1 ^ C12O + H20). The reaction therefore
goes on practically to completion. But if chlorine water is boiled
in a flask under a reflux condenser (Fig. 192), so that the distillate
constantly flows back, it is not decomposed, but remains unchanged
(Richardson, 1903). In this case the equilibrium is not disturbed,
since no constituent is removed from the sphere of action.

Chlorates. — If excess of chlorine is passed through a concentrated

solution of caustic potash
or soda, the reaction
is quite different from
that which occurs with
the cold dilute solu-
tion, described above.

EXPT. 151.— The ap-
paratus is shown in
Fig. 193. Chlorine is
generated from man-
ganese dioxide and hy-
drochloric acid in the
flask, washed with a
little water, and passed
into caustic potash
solution (20 gm. of KOH
in 40 c.c. of water) in
the beaker. Crystals
separate, and to prevent
the delivery tube be-
coming choked, an
inverted funnel is used.

When the liquid smells strongly of chlorine, it is cooled, and decanted
from the crystals which separate. If the decanted liquid is evaporated
and allowed to cool, cubic crystals separate. These, on heating with
concentrated sulphuric acid, give off fumes of hydrochloric acid : they
consist of potassium chloride. The crystals obtained by decanting the
original liquid are washed once or twice with a little cold water and
then recrystallised from hot water. They have a tabular shape
(Fig. 194), easily distinguishable from the cubes of chloride, and on
heating in a test-tube melt and evolve oxygen, leaving potassium
chloride. They consist of potassium chlorate, KC1O3.

The reaction is : 6KOH + 3C12 = 5KC1 + KC103 + 3H20.

FIG. 192.— Reflux Condenser.

XXI

THE OXIDES AND OXY-ACIDS OF CHLORINE

371

Potassium chlorate, KC103, was discovered in this way by Berthollot
in 1786; in accordance with Lavoisier's views on the nature of
chlorine (p. 221) the new salt was called hyper oxymuriate of
potash. Davy, however, showed that it was a triple compound
of potassium, chlorine, and oxygen. Potassium chlorate gives
certain reactions characteristic of all chlorates.

( 1 ) Solutions of potassium chlorate give no precipitate with silver
nitrate, but on heating

the dry salt it gives
off oxygen, and the
residue when dissolved
in water gives a
white curdy precipi-
tate of silver chloride
with silver nitrate and
dilute nitric acid :
2KC1O3=:2KC1 + 3O2;
KCl+AgN03=AgCl +
KN03

(2) If a solution of
potassium chlorate is
mixed with indigo
solution and sulphuric
acid, and a few drops
of sodium sulphite
solution are added, the
colour of the indigo
is discharged. The
chlorate is reduced by
the sulphurous acid
to a lower oxide of
chlorine, which has
strong bleaching pro-
perties.

(3) A little potass-
ium chlorate treated
with concentrated sul-
phuric acid in a test-tube turns orange -yellow, and evolves a
yellow explosive gas (chlorine dioxide, C1O2), having a peculiar odour
(p. 380). On warming there is a crackling noise, due to explosions
of the C1O2.

(4) Potassium chlorate warmed with concentrated hydrochloric acid
gives off a yellow explosive gas (euchlorine), consisting of a mixture of
Cl, and C102 : 8KC1O3 + 24HC1 = 8KC1 + 9C12 + 12H2O + 6C1O2.

B B 2

FIG. 193.— Preparation of Potassium Chlorate.

372

INORGANIC CHEMISTRY

Perchlorates. — In the decomposition of potassium chlorate by
heat, another oxy-salt of chlorine is formed, viz., potassium per-
chlorate, KC104 (p. 161) : 4KC103 = 3KC104 + KC1. This may
also be prepared by fusing potassium chlorate with barium peroxide,
extracting with hot water, and crystallising :

KC103 + Ba02 = KC1O4 + BaO.

The salt was discovered by Stadion in 1816. It is hard to say what

the oxymuriatic school would
have called it ; the chlorate
was already " hyperoxidised "
according to their views. The
crystalline form of the per-
chlorate differs from that of
the chlorate (Figs. 5 and 195).
Potassium perchlorate gives
the following reactions :

(1) It decomposes at a higher
temperature than the chlorate :
KC104 = KC1 + 202.

(2) It does not bleach indigo
in presence of sulphites.

(3) With concentrated sul-
phuric acid it does not give
a yellow explosive gas, but

dense white fumes of perchloric acid, HC1O4.
(4) It is not acted upon by hydrochloric acid.

Oxides and oxy-acids of chlorine. — By distilling hypochlorous
acid under reduced pressure, it breaks up into water and its an-
hydride, chlorine monoxide : 2HOC1 ^ C120^ +
H20. Free chloric acid is formed, in aqueous
solution, when potassium chlorate is decom-
posed by hydrofluosilicic acid, which gives
a sparingly soluble potassium salt : 2KC103
+ H2SiF6 = K2SiF6 ^ + 2HC103. By the
action of chlorine dioxide on alkali, a chloride
and a chlorite are formed : 2C1O2 + 2KOH=
KC1 + KC1O2 + H2O. Perchloric acid when

distilled under reduced pressure with phosphorus pentoxide
gives its explosive anhydride, chlorine heptoxide : 2HC104 =
H2O + C1207.

The relations between the oxygen compounds of chlorine are given
in the following table :

FIG. 194. — Crystals of Potassium Chlorate and
Chloride.

FIG. 195. — Crystal of
Potassium Perchlorate.

xxi

THE OXIDES AND OXY-ACIDS OF CHLORINE

373

Oxides.

Chlorine monoxide, or^
hypochlorous an- I C12O
hydride J

[C12O3 unknown]

Chlorine dioxide, or) pin
chlorine tetroxide J

[C12O5 unknown] ...

Chlorine heptoxide, or "I
perchloric anhydride |

Oxy- acids.
H^ocMorous acid»

Chlorous acid, HC1O2

(chlorine dioxide is a
mixed anhydride, i.e.,
one giving salts of fa;o
acids with bases)

Chloric acid, HC1O3

„ . . .

Perchloric acid, H IO4

Chlorine monoxide, C120. — This explosive substance is prepared
by distilling concentrated hypochlorous acid under reduced pressure
or by dehydrating it by the addition of fused calcium chloride :
2HOC1 ^ C100 + H9O. It is usually made by the action of dry

FIG. 196.— Preparation of Chlorine Monoxide.

chlorine on yellow precipitated oxide of mercury, previously heated
to 300-400°, contained in a cooled tube (Fig. 196). A brown
oxychloride of mercury remains, and chlorine monoxide gas
passes on :

2C12 + 2HgO = HgO,HgCl2 + C120.

It is condensed in a freezing mixture to an orange-coloured liquid,
b.-pt. 5°. The brownish -yellow gas may be collected by downward
displacement ; it attacks mercury and is soluble in water.

The gas explodes readily, although not very violently, on heating,

374 INORGANIC CHEMISTRY CHAP.

giving a mixture of two volumes of chlorine and one volume of
oxygen : 2C12O = 2C12 -f- O2. In this way its composition may be
determined. The chlorine after explosion is absorbed by caustic
soda solution. The liquid may explode if the tube containing it is
scratched with a file. If perfectly free from organic matter, how-
ever, it may be distilled without decomposition.

Hydrochloric acid is decomposed by the gas, with production of
chlorine : C12O + 2HC1 = 2C12 + H20. The gas dissolves easily
in water, forming an orange-yellow solution containing hypochlorous
acid : C120 + H2O = 2HOC1.

Hypochlorous acid, HOC1. — This acid is known only in solution.
On distillation the latter breaks up into water and the anhydride
of the acid, C12O. A solution of the acid is obtained by shaking
chlorine water with yellow precipitated mercuric oxide :

2C12 + 2HgO + H20 = HgCl2,HgO + 2HOC1.

The liquid is distilled. The anhydride passing over recombines
with water to form a dilute solution of hypochlorous acid, which
collects in the receiver.

Hypochlorous acid is most conveniently prepared from bleaching
powder, Ca'OCl'Cl.

When dissolved in water, bleaching powder is decomposed into
chloride and hypochlorite : 2Ca-OCl-Cl = CaCl2 + Ca(OCl)2. The
same solution is formed by passing chlorine through cold milk of
lime. If a clear solution of bleaching powder is treated with the
calculated amount of 5 per cent, nitric acid, added slowly from a
burette whilst the liquid is kept well stirred, hypochlorous acid is
set free : Ca(OCl)2 + 2HN03 = Ca(NO3)2 + 2HOC1. The liquid is
then*distilled, and a dilute solution of hypochlorous acid is obtained.

Any strong acid liberates from bleaching powder solution only
hypochlorous acid as a primary product, if it is not added in excess :
Ca(OCl)2 + H2SO4 = CaSO4 + 2HOC1.

Hydrochloric acid reacts with hypochlorous acid with liberation of free
chlorine : HC1 + HOC1 =± C12 + H2O. If, therefore, an excess of any
acid which is capable of liberating hydrochloric acid from calcium
chloride is added to a solution of bleaching powder, the whole of the
chlorine is expelled as such :

Ca(OCl)2 + CaCl, + 2H2SO4 = 2CaSO4 + 2H2O + 2C12.
2CaOCl2

Free hypochlorous acid is produced by the action of chlorine on
a solution of a hypochlorite, e.g., KOC1, or a solution of bleaching
powder, which contains Ca(OCl)2 :

KOC1 + C12 + H20 = KC1 + 2HOC1.

xxi THE OXIDES AND OXY-ACIDS OF CHLORINE 375

This reaction probably occurs in two stages, as follows :

(a) H20 + C12 HC1 + HOC1.
(6) KOC1 + HC1 = KC1 + HOC1.

The same result may be achieved by passing an excess of chlorine
through milk of lime or baryta water, when a hypochlorite is first
produced :

(a) Ga(OH)2 + 2C12 - Ca(OCl)2 + CaCl2 -f H20.

(6) Ca(OCl)2 + 2C12 + 2H2O = CaCl2 + 4HOC1.

If chlorine is passed through a suspension of sodium bicarbonate
or precipitated calcium carbonate in water, hypochlorous acid (not
a hypochlorite) is formed :

2C12 + H2O -f- CaC03 = 2HC1O + CaCl2 + CO2.
This reaction probably proceeds in two stages :

(i) C12 + H2O ^ HC1 + HC10 ;

(ii) HC1 + CaC03 = CaCl2 + CO2 + H2O.

The hypochlorous acid produced is too weak to decompose the
carbonate with formation of a hypochlorite. The function of the
carbonate is to remove the hydrochloric acid as fast as it is pro-
duced, and so to prevent reaction (i) coming to a standstill.

Hypochlorous acid is also produced by passing chlorine through a
solution of sodium sulphate or phosphate : Na2SO4 -f- C12 + H2O =
NaCl + NaHSO4 + HC1O. In this case the hydrochloric acid formed
as above reacts with the sodium sulphate : Na2SO4 + HC1 ^=± NaHSO4
-f- NaCl. If the liquid is distilled, however, the hypochlorous acid may
react with the sodium chloride to form sodium chlorate and free chlorine.

An interesting reaction is the formation of hypochlorous acid
by the direct oxidation of hydrochloric acid, discovered by Odling
in 1860 : HC1 -f O = HOC1. If a current of air is passed through
concentrated hydrochloric acid in a wash-bottle, and then through
potassium permanganate solution in a retort, warmed on a water-
bath, hypochlorous acid distils over.

Hypochlorous acid in solution is pale yellow, or colourless when
the solution is dilute. It is a very weak acid, practically un-ionised.
The dilute solution is fairly stable in the dark : concentrated solu-
tions decompose on heating, or exposure to sunlight, with evolution
of oxygen and chlorine, and formation of some chloric acid :

(i) 2HOC1 = 2HC1 + 02 ;
(ii) HC1 + HOC1 = H2O + 01, ;
(iii) HOC1 -f 20 (nascent) — HC103.

The decomposition is accelerated by platinum black, manganese
dioxide, or the oxides of nickel and cobalt. Hypochlorites on heat-
ing with the latter oxides in alkaline solution rapidly evolve oxygen :

376 INORGANIC CHEMISTRY CHAP.

2NaOCl = 2NaCl + O2. With concentrated acids they evolve
chlorine, as described above.

The acid dissolves magnesium with evolution of hydrogen :
Mg -f 2HOC1 = Mg(OCl)2 + H2. Iron and aluminium evolve hydro-
gen and chlorine ; copper, nickel, and cobalt evolve chlorine and
oxygen. With hydrogen peroxide the acid evolves oxygen :

HOC1 + H202 = HC1 + H20 + 02.

Hypochlorous acid is a powerful oxidising agent. Its bleaching
action is due to the liberation of nascent oxygen : HOC1 = HC1 + O.

EXPT. 152. — Add caustic soda to a solution of manganous sulphate.
A white precipitate of manganous hydroxide is formed: MnSO4 +
2NaOH = Mn(OH)2 + Na2SO4. Add sodium hypochlorite solution.
The precipitate instantly becomes brown, and is converted into hydrated
manganic oxide: Mn(OH)2 + NaOCl + H2O = Mn(OH)4 + NaCl.

EXPT. 153. — To a solution of chrome alum add caustic soda ; a green
precipitate of chromic hydroxide, Cr(OH)3, is formed. Add excess of
NaOCl solution and NaOH, and boil. A yellow solution of sodium
chromate, Na2CrO4, is formed : 2Cr(OH)3 + SNaOCl + 4NaOH =
2Na2CrO4 + 5H2O + 3NaCl.

Bleaching powder. — Chlorine gas does not react with quicklime
at the ordinary temperature, but at a red heat oxygen is expelled
and calcium chloride formed : 2CaO + 2C12 = 2CaCl2 + O2. If,
however, chlorine is passed over dry slaked lime, Ca(OH)2, it is
rapidly absorbed, forming a somewhat moist powder which smells
of hypochlorous acid, and is called bleaching powder, or chloride of
lime. The reaction is : Ca(OH)2 -f C12 = CaOCl2 + H2O, the water
formed remaining principally in the powder.

In the manufacture of bleaching powder the slaked lime is spread
over the floors of closed lead chambers, so as to expose a large
surface, and somewhat diluted chlorine gas admitted. At first the
chlorine is rapidly absorbed by the lime, but the reaction afterwards
slows down. The powder is then turned over with wooden rakes,
and the action of the gas continued until absorption is complete,
which takes 12-14 hours. The product usually contains 37-39 per
cent, of chlorine present as CaOCl2, whereas that calculated from
the formula CaOCl2 -f- H2O is 49. Some free lime is also present.

With very dilute chlorine, such as is produced by the Deacon
process, it is necessary to provide a very intimate contact of the
lime with the gas. This is effected by making the gas traverse
lead or iron pipes placed horizontally one above the other, through
which the lime is pushed in the opposite direction to the gas
by means of Archimedean screws (Hasenclever screw - chambers,
Fig. 197). The lime drops from one pipe to the other and is
withdrawn into casks at the bottom fully charged with chlorine.

THE OXIDES AND OXY-ACIDS OF CHLORINE

377

The formula of bleaching powder. — Bleaching powder was at first
regarded as a molecular compound of lime and chlorine — " chloride
of lime," CaO,Cl2. Balard in 1835 suggested that it was a mixture
of equimolecular amounts of calcium hypochlorite and chloride :

rrnnnnnnnnn n-n

\_yu_ u u u u u u

FIG. 197. — Hasenclever Bleaching Powder Apparatus.

Ca(OCl)2 -f- CaCl2. Commercial bleaching powder always contains
an excess of free lime, which led Stahlschmidt to assume that it con-

tained the compound Ca/ , formed according to the equation :

XOC1
,OR

2C12 = 2Ca

CaCl2 + 2H2O.

Later experiments showed, however, that free lime is not an essential
constituent, but is merely due to the particles of lime becoming
encrusted with bleaching powder, and so escaping complete chlorina-
tion. The reaction appears to be :

Ca(OH)2 + C12 - [CaOCl2 + H20].
Lunge prepared a product of the following composition :

CaOCl2,H2O = 91-80
CaCO3 0-95

CaCL
Ca(OH)2

0-45
6-80

100-00

378 INORGANIC CHEMISTRY CHAP.

Balard's formula, Ca(OCl)2 -f CaCl2, would require that bleaching
powder should contain a considerable proportion of free calcium
chloride. If, however, it is treated with successive small amounts
of water, the first portions of the extract contain much less
chlorine as chloride than would be the case if the latter pre-existed
in the powder. Again, alcohol extracts from good bleaching powder
only a small amount of calcium chloride, although the latter is
readily soluble in that solvent.

These results agree with the formula proposed by Odling, accord-
ing to which the active constituent of bleaching powder is a mixed

,OCl
salt of the formula Ca^ , i.e., calcium chloro-hypochlorite, formed

by the simultaneous neutralisation of a molecule each of hydro-
chloric and hypochlorous acids :

/OH HOC1 /OC1 H20

Ca<^ + = Car

HC1 C1 H20

Stahlschmidt's formula for the active constituent of bleaching
powder, Ca'OH'OCl, is disproved by the fact that, although bleaching
powder containing as much as 48-74 per cent, of chlorine which can
be liberated by acids, i.e., available chlorine, has been prepared, his
formula limits this to 33 per cent.

O'Shea (1883) decided between the three rival formulae :

Balard's Ca(OCl)2 + CaCl2,
Stahlschmidt's Ca.OH.OCl,
Odling's Ca.OCl.Cl,

as follows. He removed any free calcium chloride by treatment
with alcohol, and determined in the residue : (i) the total lime,
CaO ; (ii) the total chlorine ; (iii) the chlorine as hypochlorite. The
following ratios were found :

lime : total chlorine = 1:2; lime : hypochlorite chlorine =1:1;
hypochlorite chlorine : total chlorine = 1:2.

The residue after treatment with alcohol, and the above ratios,
should be, in the different cases :

CaO QaO hypochlorite Cl.

Residue. total Cl hypochlorite Cl total Cl

1. Balard ... Ca(OCl)2 1:2 1:2 1:1

2. Stahlschmidt ... Ca.OH.OCl 1:1 1:1 1:1

3. Odling ... Ca.OCl.Cl 1:2 1:1 1:2

Thus, only Odling's formula agrees with the experimental results.

xxi THE OXIDES AND OXY-ACIDS OF CHLORINE 379

Available chlorine of bleaching powder. — Bleaching powder is
mainly employed as an oxidising agent, and the active agent is
really the nascent oxygen of the hypochlorite. Usually, however,
the chlorine equivalent of this active oxygen is returned as the
available chlorine : O (16) = C\2 (71). If the bleaching powder
consisted entirely of the compound Ca'OCl'Cl, the chlorine equivalent
of the active oxygen atom of the hypochlorite radical would be
O = C12, i.e., the total chlorine in the compound. This would, in fact,
be wholly expelled by acids : CaOCl2 + H2SO4 = CaSO4 + H2O +
C12, in accordance with the former definition of available chlorine.

As it is met with in commerce, however, bleaching powder always
contains some free calcium chloride, CaCl2, and possibly calcium
chlorate, Ca(C103)2, and since the chlorine of these compounds is not
liberated as such by acids, and the oxygen of the chlorate is not
available for the usual oxidising purposes of bleaching powder, a
distinction is made between the total and available chlorine.

The estimation of the available chlorine of bleaching powder is
carried out by one of the following methods :

1. Penot's method : About 10 gm. of bleaching powder are weighed
out into a mortar, and triturated with successive quantities of cold
distilled water until the paste has been transferred to a litre flask, which
is filled to the mark with water, and well shaken. 50 c.c. of the well-
shaken suspension (a little powder remains undissolved) are now pipetted
into a beaker, and titrated with decinormal sodium arsenite solution,
until a drop of the liquid, placed by means of a glass rod on a piece of
filter-paper which has been soaked in potassium iodide and starch solution
and dried, no longer gives a blue colour owing to liberation of iodine :
2KI + C12 = 2KC1 + I2. The reaction is : As2O3 + 2CaOCl2 =
As2O5 -f- 2CaCl2. Thus As2O3 requires 2O or 4C1 (4C1 + 2H2O =

4HC1 + 2O), so that 1 c.c. of ~^As2O3 == 0-00352 gm. of active Cl.

The decinormal sodium arsenite, Na3AsO3, is prepared by dissolving
4-95 gm. of pure arsenious oxide, As2O3, and 25 gm. of pure sodium
bicarbonate, NaHCO3, in 500 c.c. of distilled water in a flask heated
on a water-bath, cooling, and diluting to 1 litre in a measuring flask.

2. Bunsen and Wagner's method : 25 c.c. of the bleaching powder
suspension are treated with excess of potassium iodide solution, and
acidified with acetic acid. Iodine is liberated : 2KI -f HOC1 -f-
CHg-COOH = 2CH3-COOK (potassium acetate) + I2 + H2O. This is
titrated with decinormal sodium thiosulphate solution until the yellow
colour has practically vanished : 2Na2S2O3 + I2 = Na2S4Oc (sodium
tetrathionate) + 2KI. A little starch-paste is then added, and the
titration continued until the blue colour, due to the iodine, vanishes.

N/W Na2S2O3 solution contains 24-8 gm. of Na2S2O3,5H2O per litre ;

380 INORGANIC CHEMISTRY CHAP.

it is standardised by JV/10-iodine solution. 1 c.c. = 0-00352 gm. of
active Cl.

Hypochlorous acid, or hypochlorites, are estimated in presence of
free chlorine by means of the following reactions :

2KI + HOC1 + HC1 = 2KC1 + I2 + H2O
2KI + C12 = 2KC1 + I2.

Each molecule of HOC1 neutralises one equivalent of acid, whilst
chlorine does not affect the acidity of the solution. By titrating the
iodine and the remaining acid, the amounts of HOC1 and C12 may be
calculated.

Hypochlorites. — A solution of sodium hypochlorite is prepared
by methods previously described (p. 368). It is used in America
for bleaching purposes instead of bleaching powder, and then con-
tains 1 to 2 per cent, of NaOCl. By cooling a concentrated solution,
from which sodium chloride has deposited, to — 10°, and shaking,
crystals of NaOCl,6H20, or NaOCl,7H2O, separate. These are
very deliquescent, and melt at 18°. On cooling the fused substance,
large crystals of NaOCl, 5H2O are formed.

Calcium hypochlorite, Ca(OCl)2, is prepared in crystals by passing
chlorine through milk of lime, and evaporating the clear solution in
vacuo. It is more stable than bleaching powder, is completely
soluble in water, and contains 80-90 per cent, available chlorine
(theoretical hypochlorite oxygen in Ca(OCl)2 = 224 per cent.,
hence equivalent of Cl = 224 x 704/16 = 98-5 per cent.). The
crystalline hydrate, Ca(OCl)2,4H2O, is first deposited on evaporation.

Chlorine dioxide, C102 and C1204. — By the action of concentrated
sulphuric acid on potassium chlorate Davy (1815) obtained a yellow
explosive gas. On explosion, two volumes of this gas gave three
volumes of gas, consisting of two volumes of oxygen and one volume
of chlorine, hence its formula is C1O2:2C102 = C12 + 202.

Powdered, previously fused, potassium chlorate is added in small
quantities at a time to cooled concentrated sulphuric acid in a small
retort. The orange-yellow paste is very cautiously warmed by placing
the retort in lukewarm water, and the gas collected by downward dis-
placement, since it is heavier than air. It dissolves in water and attacks
mercury. There is considerable danger of violent explosion in the pre-
paration of chlorine dioxide.

The reaction appears to take place as follows :

KC103 + H2S04 = KHS04 + HC103 (chloric acid) ;
3HC1O3 = HC1O4 (perchloric acid) + 2C1O2 -f H2O.

The density of chlorine dioxide gas was found by Pebal and
Schacherl (1882) to correspond with the formula C1O2. If passed
through a tube cooled in a freezing mixture the gas condenses to a

xxi THE OXIDES AND OXY-ACIDS OF CHLORINE 381

dark red liquid, boiling at 9°, and at — 79° this freezes to an orange-
cole^ ured crystalline solid. The liquid and solid are believed to
have the formula C12O4. The liquid is violently explosive, although
it may be distilled without decomposition in the entire absence of
organic matter. The gas also explodes readily on heating with a hot
wire or glass rod, by an electric spark, or in contact with turpentine,
alcohol, or ether.

EXPT. 154. — Add a few c.c. of cold concentrated sulphuric acid to
two portions of 1 gm. of potassium chlorate in two test-tubes. A
yellow gas with a peculiar smell is generated. Insert a hot glass rod
into one tube ; into the other throw a small piece of phosphorus. The
gas in the first tube explodes ; the phosphorus in the second tube in-
flames spontaneously and explodes the gas.

Chlorine dioxide (sometimes called chlorine peroxide) is a powerful
oxidising agent. This is evident from some of the following experi-
ments.

EXPT. 155. — Equal parts of powdered sugar (or starch) and potassium
chlorate are mixed with a spatula on a piece of paper, and a drop of
concentrated sulphuric acid is allowed to fall on the mixture from a glass
rod. The whole mass ignites, and burns violently.

EXPT. 156. — A little potassium chlorate is placed in a glass of water,
and one or two small fragments of phosphorus are thrown in. If a few c.c.
of concentrated sulphuric acid are poured carefully down a thistles
funnel on to the chlorate, C1O2 is evolved. When the bubbles of this
gas come in contact with the phosphorus, there is a series of flashes of
light, accompanied by slight and harmless explosions.

EXPT. 157. — Chlorine dioxide, generated from 1 gm. of previously
fused chlorate and 3 c.c. of cold concentrated sulphuric acid in a small
jar, is decanted into a second jar containing leaves of Dutch metal.
No action occurs. The gas is then exploded by a hot wire, when the
metal at once takes fire in the chlorine liberated.

EXPT. 158. — A drop of a solution of phosphorus in carbon disulphide
is allowed to fall on a small quantity of powdered potassium chloride.
When the carbon disulphide has evaporated there is a sharp explosion.

Chlorous acid, HC102. — Chlorine dioxide dissolves in water, form-
ing a yellow solution without acid reaction. With solutions of
alkalies, however, it acts as an acidic oxide, forming a mixture of two
salts in equivalent amounts : 2KOH + 2C102 = KC103 + KC102 -f
H20. It is a mixed anhydride, forming the salts of two acids with
bases. KC102 is the salt of chlorous acid, HC1O2. The two salts
may be separated by evaporation in vacuo over sulphuric acid,
when the less soluble KC103 is first deposited.

382 INORGANIC CHEMISTRY CHAP.

Pure chlorites may be obtained by the action of alkali and hydro-
gen peroxide on a concentrated aqueous solution of chlorine dioxide.
The latter is conveniently prepared by warming to 60° a mixture of
40 gm. of KC1O3, 150 gm. of crystalline oxalic acid, and 20 c.c. of
water, and passing the gas into water : 2KC103 -f 2C2H2O4 =
K2C2O4 + 2H2O + 2C02 + 2C102. When diluted with carbon
dioxide, chlorine dioxide is not liable to explode during preparation.
The hydrogen peroxide reduces chlorine dioxide to chlorous acid :
2C102 -f H202 = 2HC102 + 02.

The alkali chlorites have a ' caustic taste, and bleach vegetable
colours. They may be distinguished from hypochlorites by the
bleaching action after addition of sodium arsenite. Silver and lead
nitrates precipitate yellow crystalline AgClO2 and Pb(C102)2. These
explode on heating ; lead chlorite detonates violently on per-
cussion when mixed with sugar, and is used for detonators.
Free chlorous acid is obtained in solution by treating a chlorite
with oxalic acid.

The anhydride of chlorous acid would be C1203, but is not known.
The gas prepared by heating a mixture of potassium chlorate and
sugar, benzene, or arsenious oxide, with nitric acid, believed to be
the trioxide of chlorine by Millon (1845), was shown by Garzarolli-
Thurnlack and Schacherl to be a mixture of chlorine dioxide with
chlorine. The mixture of chlorine and the dioxide obtained by
treating potassium chlorate with concentrated hydrochloric acid,
supposed by Davy to be an oxide of chlorine, and called euchlorine,
was examined by Pebal, and the method used was applied by the
above experimenters to the supposed trioxide.

A measured volume of the gas was exploded by heating, and the
increase in volume determined. The chlorine was then absorbed
by potash solution, and the residual oxygen measured. A com-
parison of the expansion on explosion with the volume of oxygen
produced showed that the latter was double the former.

The different oxides of chlorine would give the following
results :

Residual oxygen

Expansion on after absorption

Explosion. of chlorine.

(1) Chlorine monoxide : C12O

2C12O = 2C12 + O2 3-2 = 1 vol. 1 vol.

(2) Chlorine dioxide, C1O2

2C1O2 = C12 + 2O2 3 — 2 = 1 vol. 2 vols.

(3) Chlorous anhydride, C12O3 (?)

2C12O3 = 2C12 + 3O2 ... 5 — 2 = 3 vols. 3 vols.

With euchlorine and the supposed trioxide, the volume relations
(2) were found, hence both contained only chlorine and chlorine dioxide.
By passing the " trioxide," and euchlorine, through tubes immersed

xxi THE OXIDES AND OXY-ACIDS OF CHLORINE 383

in a freezing mixture, pure chlorine dioxide was liquefied, and chlorine
passed on.

Chloric acid, HC103. — Chloric acid is much more stable than
hypochlorous acid ; it is formed when the latter, or chlorine water,
is exposed to light. If a solution of potassium chlorate is precipi-
tated with hydrofluosilicic acid, sparingly soluble potassium silico-
fluoride is formed, and the aqueous chloric acid can be filtered off :
2KC103 -f H2SiF6 = K2SiF6 -f 2HC103. It is most convenient
to start with barium chlorate, a solution of which is precipitated with
the calculated amount of sulphuric acid : Ba(C103)2 + H2S04 =
BaS04 (ppd.) + 2HC103. The solution is decanted from the barium
sulphate, and evaporated in a vacuum desiccator over concentrated
sulphuric acid until it contains 40 per cent of HC103. On further
concentration the acid decomposes into chlorine, oxygen, and per-
chloric acid.

Barium chlorate is made by evaporating a solution of sodium
chlorate and barium chloride : 2NaC103 -f BaCl2 ^ 2NaCl +
Ba(C103)2. The sodium chloride is deposited, and is fished out :
the remaining solution is crystallised. Chloric acid was prepared
from barium chlorate by Gay-Lussac in 1814. The concentrated
acid is colourless, and fairly stable in the dark. When exposed
to light it undergoes decomposition (see above) and becomes yellow.
Organic substances, such as cotton-wool or paper, are ignited by
the concentrated acid. It has a pungent smell, and strong acidic
and bleaching properties. The anhydride corresponding with
chloric acid, 2HC103 — H20 — C1205, is not known.

EXPT. 159. — Pour a concentrated solution of sodium hydrogen sulphite
(NaHSO3) over crystals of potassium chlorate. A trace of free chloric
acid is liberated by the weakly acid NaHSO3. The latter is then
oxidised by the chloric acid to the strongly acid NaHSO4. More chloric
acid is liberated, and the velocity of reaction is increased by the action
of the products (i.e., by autocatalysis) until in one or two minutes the
whole mixture foams over, acid sodium sulphate (NaHSO4) and
hydrochloric acid being formed.

Chloric acid is ionised in solution, and is a strong acid : HC103 ^
H' -f- C10'3. In acid solutions chlorates are readily reduced by iron
or aluminium powder to hydrochloric acid. In this way they may
be estimated : HC103 + 3H2 = 3H2O + HC1.

Perchloric acid, HC104. — The most stable oxy-acid of chlorine
is that containing most oxygen, viz., perchloric acid, HC104.
Small quantities of the very soluble sodium salt, NaC104, occur in
Chile nitre (p. 563) : they act prejudicially on vegetation if the impure
nitrate is used as a fertiliser.

Perchloric acid is formed by the evaporation of a solution of

384 INORGANIC CHEMISTRY CHAP.

chloric acid, and if the latter is distilled, aqueous perchloric acid
comes over, chlorine and oxygen escaping at the same time.

Potassium perchlorate, KC1O4, is prepared by heating the chlorate
at about 400 ° until it becomes pasty, and separating from the more
soluble chloride by crystallising from hot water : 2KC103 = KC104-f-
KC1 + O2, or 4KC1O3 = 3KC104 + KC1. Any chlorate remaining
may be decomposed by hydrochloric acid, which is without action on
the perchlorate. If potassium perchlorate is distilled with four times
its weight of very concentrated sulphuric acid in a small retort, per-
chloric acid comes over as a colourless, or slightly yellow, strongly
fuming liquid : KC1O4 + H2SO4 = KHS04 -f HC1O4. The yield
is increased by carrying out the distillation under 10-20 mm.
pressure, when the acid distils over between 90° and 160°. It is
purified by distilling under 60 mm. pressure, when it boils at 40-60°.
It boils, with partial decomposition, at 90° under 760 mm. pressure,
or without decomposition at 19° under 11 mm. pressure.

During the distillation under ordinary pressure, the liquid in the
receiver gradually solidifies to white crystals of the monohydrate,
HC104,H2O, m.-pt. 50°. Other crystalline hydrates are known :
HC104,2H20 (m.-pt. - 20-6°) ; 2HC104,5H2O (m.-pt. - 30°) ; and
two forms of HC104,3H2O (m.-pt. — 43-2° and -37°). The
anhydrous acid is very hygroscopic, and dissolves in water with
a hissing noise and great evolution of heat. The hydrate HC10 4,H20
was regarded as the acid itself by its discoverer, Stadion (1816) ;
pure HC104 was first prepared by Roscoe (1863). On heating,
HC104,H2O 'breaks up into anhydrous acid, which distils over, and
an oily solution of maximum boiling point, 203°, containing 72
per cent, of HC1O4.

The oily aqueous acid, which is quite stable, is conveniently pre-
pared by adding ammonium perchlorate (a commercial substance), dis-
solved in concentrated hydrochloric acid, to warm concentrated nitric
acid in a porcelain dish. Nitrogen, chlorine, and nitrosyl chloride
are evolved, and on evaporation aqueous perchloric acid remains :

NH4C104 + HC1 - NH4C1 + HC104
HN03 + 3HC1 = NOC1 + C12 + 2H20
2NH4C1 + 3C12 = N2 + 8HC1.

Anhydrous perchloric acid is liable to explode spontaneously ;
paper and wood catch fire when it is dropped on them. If a few
drops of the acid are poured on recently -ignited wood charcoal, there
is a violent explosion.

The aqueous acid dissolves iron and zinc to form perchlorates .
2HC104 + Zn = Zn(ClO4)2 + H2, and the acid is not reduced
(cf. HC103). It is reduced only by sodium hyposulphite (Na2S204),
titanium trichloride, or, in alkaline solution, by ferrous hydroxide.
It is therefore a much less powerful oxidising agent than chloric acid.

xxi THE OXIDES AND OXY-ACIDS OF CHLORINE 385

The preparation of acids, — The preparations of perchloric and
chloric acids illustrate two general methods for the preparation of
acids. A salt of the acid is acted upon by another acid. A state
of equilibrium then results, in which all four compounds are present
EX -f HA ^-RA + HX. It may not be possible in practice
to separate the acid HX from the other substances, but if it can be
separated, the equilibrium is disturbed, and the reaction may go
on nearly to completion. . Separation is possible when the acid HX is
volatile, as in the case of perchloric acid. It is then distilled off.
If the acid is not appreciably volatile, as in the case of chloric acid,
a second method may be used. In this, the salt RX and the acid
HA are so chosen that the salt RA is practically insoluble ; it is then
filtered off, and the acid HX is left in solution.

Thus R may be Ba" or K>, and A, SO4" or SiF6" respectively,
since BaS04 and K2SiF6 are only sparingly soluble in water. It will
be found that nearly all the methods described for the preparation
of acids from their salts are special cases of these two general
methods.

Chlorine heptoxide, C1207. — The anhydride of perchloric acid,
C12O7, was discovered by Michael and Conn in 1900. 10 gm. of
phosphorus pentoxide are placed in a small stoppered retort con-
nected with a phosphorus pentoxide drying-tube and a receiver
cooled in ice and salt. Pure perchloric acid is added, in quantities
of 10 drops at a time, and allowed to trickle down the sides of the
retort on to the P205 : an interval of ten minutes is allowed to
elapse after each addition, and the retort is kept at a temperature of
- 10° in a freezing mixture. After allowing to stand twenty-four
hours in the freezing mixture, the retort is warmed to 85°, and a
colourless oily liquid distils over, boiling at 82°. This is perchloric
anhydride, C1207 : 2HC104— H20 = C1207. Violent explosions
may occur in its preparation, although C1207 is more stable than
C12O or C102, and may be poured on paper, wood, sulphur, or
phosphorus, without explosion. It explodes when heated or struck,
and decomposes on standing for a few days. It sinks in water, and
slowly forms HC104 : C1207 + H20 = 2HC104.

The manufacture of chlorates and perchlorates. — Chlorates are
manufactured either by the action of excess of chlorine on
concentrated solutions of alkalies, or by the electrolysis of chlorides.
Calcium chlorate is produced by passing chlorine into hot milk of
lime contained in cast-iron vats, with agitating paddles (Fig. 198),
until the reaction is complete. Lunge and Landolt represent the
'reaction as follows :

(1) 2Ca(OH)2 + 2C12 *= Ca(OCl)2 + CaCl2 + H20.

(2) Ca(OCl)2 + 2C12 + 2H20 - CaCla + 4HC1O.
(3)-2Ca(OCl)2 + 4HC10 = CaCl2 + Ca(C103)2 + 2C12 + 2HaO.

c c

386 INORGANIC CHEMISTRY CHAP.

The complete reaction is : 6Ca(OH)2 + 6C12 = 5CaCl2 + Ca(ClO3)2
+ 6H20, but this appears to take place with the intermediate
formation of hypochlorous acid, which acts as a carrier of oxygen.
The action of heat alone on calcium hypochlorite, in the absence of
excess of chlorine, is mainly according to the equation : Ca(OCl)2 =
CaCl2 -f- O2. Alkaline hypocnlorite solutions may be boiled without
much decomposition, but oxygen is slowly evolved. Traces of
chlorites are also formed.

The solution of calcium chlorate may be treated with potassium
chloride, when the sparingly soluble . potassium chlorate crystallises
out, and is recrystallised. It is now usual to produce the very
soluble sodium chlorate, NaC103. The solution of calcium salts
is concentrated, cooled, and filtered from the crystals of hydrated
calcium chloride which separate. Excess of sodium sulphate is

then added, when all the calcium
is precipitated as. sulphate. On
evaporation of the filtered solu-
tion, sodium chloride separates ;
this is removed, and, on cooling,
sodium chlorate crystallises out.
Chlorates and perchlorates are
also produced by the electrolysis
of saturated sodium chloride solu-
tion at 80°, between platinum
electrodes placed close together.
A little potassium chromate is
added as a catalyst. The
chloride is first completely
converted into chlorate ; on pro-
longed electrolysis, this passes
into perchlorate. There are large
chlorate works in Switzerland and at Niagara. Chlorates are used
as oxidising agents (e.g., in the oxidation of aniline to aniline
black), and in making fireworks. Perchlorates are employed in the
manufacture of detonators and explosives.

Heat of reaction. — The evolution of heat which accompanies large
numbers of chemical reactions, in some cases appearing as active com-
bustion, is of great importance in technical processes. The greater
part of the energy expended in the affairs of daily life proceeds from
the combustion of coal, in other words from a chemical process. The
value of coal is not in its chemical constituents, since these are
almost entirely dissipated and lost in the ashes and flue-gases during
its combustion, but is determined by the amount of energy in the
form of heat which can be obtained by the combustion of the fuel.

It follows from the Law of Conservation of Energy that the energy
contained in the unburnt coal, and in the oxygen of the air, must

FIG. 198.— Manufacture of Calcium
Chlorate.

xxi THE OXIDES AND OXY-ACIDS OF CHLORINE 387

exceed that in the ash, and in the gaseous products of combustion,
by the amount of heat evolved. The latter is thus a measure of the
difference between these two stores of energy.

The stores, or charges, of chemical energy associated with material
systems may be set free in the form of heat (or electrical energy,
p. 879) during chemical reactions. In these changes, however, only
a portion of the energy of the materials is set free ; another part
remains associated with the products. How much energy is associated
with matter we have no certain means of judging ; all that can be
determined is the difference between the energies of the systems
before and after change has occurred. This difference is evolved
as heat, and may be measured.

If the reacting matter is in the gaseous form, considerable changes
of volume may occur, and hence work is done by the pressure of the
atmosphere on the system, if there is a contraction, or is spent by the
system in overcoming that pressure, if there is an expansion. In
the former case, the evolution of heat is greater, by the thermal
equivalent of the external work, than it would have been if no change
of volume had occurred. In the latter case, the heat evolved is
diminished by that part of the energy of the system which would
otherwise have appeared as heat, but now leaves the system as
external work spent in overcoming pressure. To obtain the net
diminution of energy of the system, the reaction must be carried out
at constant volume ; or, if the volume changes, a correction for
the external work must be applied to the heat evolved. If there is
contraction, work is spent on the system and appears as heat, so that
this must be subtracted from the total heat evolved. If there is
expansion, part of the energy appears as work, and its equivalent in
heat must be added to the observed heat evolution.

We have therefore to distinguish between heats of reaction at
constant volume and heats of reaction at constant pressure.

A mixture of 2 gm. of H2 and 16 gm. of O2 at 0° and 1 atm. pressure
occupies 22,240 + 11,120 = 33,360 c.c. If this is converted into liquid
water at 0°, the latter will occupy 18 c.c. There has been a diminution
in volume of 33,360 — 18 = 33,342 c.c., and since the atmospheric pres-
sure is equal to 76 X 13*6 X 981 dynes per sq. cm. (p. 149), the work done
by the atmospheric pressure on the system, which appears as heat, is
33,342 X 76 x 13'6 X 981 = 3-38 X 1010 ergs = 3-38 X 1010/4-18 X
107 gm. cal. = 808 '5 gm. cal. The observed heat of reaction at constant
pressure is 68,939 gm. cal., hence the heat of reaction at constant volume
is 68,939 — 808-5 = 68,130 gm. cal. This latter value represents the
difference between the chemical energies of the hydrogen and oxygen
gases, and that of the liquid water. Thus :

2H -f O = H2O (liq.) + 68,939 gm. cal. (constant pressure)
2H + O = H2O (liq.) + 68,130 gm. cal. (constant volume).

C C 2

388 INORGANIC CHEMISTRY CHAP.

If the reaction occurred at 100°, with production of steam, the heat
evolved is diminished by the latent heat of steam, 18 X 538 gm. cal.

Hess's law. — If a reaction is carried out in stages, the algebraic sum of
the amounts of heat evolved in the separate stages (heat absorbed being
reckoned negative) is equal to the total evolution of heat when the
reaction occurs directly.

This simple consequence of the Law of Conservation of Energy
is known as Hess's Law* (1840). It enables one to calculate many
heats of reaction which could not be determined directly.

EXAMPLE 1. — Find the heat of formation of carbon monoxide, CO,
from solid carbon and gaseous oxygen, given the following data :

Heat of combustion of carbon to carbon dioxide : C + 2O = CO2

+ 97 kgm. cal.
Heat of combustion of carbon monoxide to dioxide : CO + O

= CO2 + 68 kgm. cal.

By subtracting the second of these equations from the first, we find :
Heat of formation of carbon monoxide : C + O = CO + 29 kgm. cal.

EXAMPLE 2. — Find the heat of formation of ammonia from its
elements, given :

(1) 2NH3 + 3O = 3H2O (liq.) + N2 + 18,120 gm. cal.

(2) 2H + O = H2O (liq.) + 68,939 gm. cal.
Multiply (2) by 3, and subtract from (1) :

2NH3-f 3O - 6H - 3O = 3H2O + N2- 3H2O + 18,120 -206,871.
/. N2 + 3H2 = 2NH3 + 188,697 gm. cal. ;
.-. N + 3H = NH3 -f 94,348 gm. cal.
Heats of formation are calculated for 1 gm. mol. of the compound.

Thermochemistry. — That branch of chemistry which is concerned
with heats of reaction is called thermochemistry. The fundamental
law is that of Hess, and by means of this all heats of reaction may be
calculated from the heats of formation of the compounds con-
cerned. These heats of formation have been determined experiment-
ally, chiefly by Julius Thomsen, and by Marcellin Berthelot, and are
tabulated per gm. mol. of compound produced.

If we suppose all the compounds on the left of an equation to be
decomposed into their elements, an amount of heat is absorbed equal
to the algebraic sum of the heats of formation of these compounds.
If we now suppose the elements to be combined to form the com-
pounds on the right of the equation, an amount of heat is evolved
equal to the algebraic sum of the heats of formation of these com-
pounds. It follows from Hess's law that :

Heat of reaction = Heat of formation of final compounds — Heat of
formation of initial compounds.

The energies of the compounds are all referred to those of the

,he

xxi THE OXIDES AND OXY-ACIDS OF CHLORINE 389

elements as zero. The amounts of energy associated with the
different elements are not, of course, zero, nor are they equal, but
it is only the difference between the amounts of energy associated
with the elements when in combination and when free that is
required.

Thus, the equation Cu -f C12 = CuCl2 + 51 -6 kg. cal. may be
written in the form: 0 = CuCl2-f-51-6 kg. cal., or CuCl2 =
— 51 -6 kg. cal., indicating that CuCl2 contains 51 -6 kg. cal. less energy
than Cu -f- C12. The symbols of compounds thus represent quantities
of energy, which may be added or subtracted. We may therefore, in
the thermochemical equation, write the negative values of the heats
of formation instead of the chemical symbols, and solve for the
unknown heat of reaction.

EXAMPLE 3. — Find the heat of the reaction :

CaCl8 + 2Na - Ca + 2NaCl + x.

The heats of formation of CaCl2 and NaCl are 170 kg. cal. and 97-8
kg. cal., respectively, hence :

- 170 = — 2 x 97-8 + ay, or x = 25'6 kg. cal.

If substances are produced in aqueous solution, we have to take
account of the heats of solution. These vary with the amount of
water, but become constant when this is very large ; we usually
suppose so much water taken that the heat of solution is constant.
This amount of water is denoted by Aq. Thus : NH3 (gas) -f- Aq =
NH3,Aq 4- 8400 gm. cal. means that when 17 grams of ammonia
gas dissolve in a large quantity of water, 8400 gm. cal. are evolved.
If still more water is added, no heat change occurs, hence Aq does
not need to be specially stated. The heat of solution of perchloric
acid is very large : HC104 + Aq = HC104,Aq -f 20,100 gm. cal.

The stability of compounds. — We have frequently used the terms
stable and unstable to denote whether a given compound is with
difficulty resolved into its elements, or into related compounds,
or whether this change takes place easily and spontaneously. Thus,
water and hydrogen chloride are stable compounds : they show no
tendency to decompose spontaneously into their elements, or into
other compounds of these. The oxides and oxy-acids of chlorine, on
the other hand, are all unstable substances, decomposing spon-
taneously, or when heated, or when brought in contact with other
substances. We have also seen that there are different degrees of
stability ; thus the stability of perchloric acid is greater than that of
hypochlorous acid.

It is of interest to inquire into the causes of the stability (or other-
wise) of substances. Formerly only certain empirical rules, which
had numerous exceptions, were available for this inquiry.

Thus, the stability of a compound depends on the electrochemical

390 INORGANIC CHEMISTRY CHAP.

character (and therefore on the position in the Periodic System, cf. p. 455)
of its component elements. Compounds of strongly electropositive with

+' -
strongly electronegative elements are usually stable, e.g., KC1 ; whilst

compounds of elements of the same electrochemical character are usually

+ +
unstable, e.g., C12O, Pd2H. An exception is the very stable P2O5.

Again, the stability of a compound alters with the valency of an element

IV II

contained in it. Thus PtCl4 decomposes at 300° into PtCl2 and C12 ;
II
PtCl2 is decomposed only at 500°, and is therefore more stable. The

atomic linkage also affects the stability ; in carbon compounds, those
which are saturated (single linkages) are much more stable than those
which are unsaturated, or contain double or treble bonds (p. 250).
Thus, ethane, CH3-CH3, is quite stable, whereas acetylene, CHjCH,
is explosive.

The principal condition affecting the stability of a substance,
however, is the quantity of energy it contains.
If we examine the thermochemical equations :

2H + O = H20 (liq.) + 684 kg. cal.
H + Cl = HC1 + 22 kg. cal.
K + Cl = KC1 + 106 kg. cal.
C + 2H2 = CH4 -f 21-8 kg. cal.

we see that all the above compounds, water, hydrochloric acid,
potassium chloride, and methane, are formed from their elements
with considerable evolution of heat, i.e.. loss of energy — i.e., they are
strongly exothermic compounds. They contain considerably less
energy than the elements from which they are produced, and their
properties show that they are stable.

Now consider the following thermochemical equations :

2C1 + 0 = C120 - 17-8 kg. cal.
HCl,Aq+ 0 = HOd,Aq — 9-3 kg. cal.
HCl,Aq+ SO = HC103,Aq - 15 kg. cal.
HCl,Aq+40 = HC104,Aq- 0-7 kg. cal.

The compounds chlorine monoxide, hypochlorous acid, chloric
acid, and perchloric acid are formed from the substances on the left
with absorption of heat ; they are endothermic compounds, and contain
more energy than their constituents. They are all unstable, and
tend to decompose.

In general, a substance formed with considerable evolution of
energy will be stable, whilst a compound formed with considerable
absorption of energy will be unstable. The instability is roughly
in proportion to the amount of energy absorbed in formation ; thus,
perchloric acid is more stable than either hypochlorous acid or chloric

xxi THE OXIDES AND OXY- ACIDS OF CHLORINE 391

acid, although hypochlorous acid is less stable than the other two.
Aqueous perchloric acid is formed with considerable evolution of
heat, and is quite stable.

The constitution of the oxy-compounds of chlorine. — If we assumed
chlorine to be univalent in all its oxygen compounds, these would
have the following formula) :

chlorine monoxide, Cl — O — Cl hypochlorous acid, H— O — Cl
chlorine dioxide, Cl — O — O — Cl chlorous acid, H— O — O — Cl
chloric acid, H — O — O — O — Cl perchloric acid, H — O— O — O — O — Cl
chlorine heptoxide, Cl— O— O— O— O— O— O— O— Cl

It is usually noticed, however, that the stability of compounds
containing chains of singly-linked oxygen atoms decreases as the
number of oxygen atoms in the chain increases. Thus,, hydrogen
peroxide, H — 0 — 0 — H, is less stable than water H— O — H. We
should therefore expect the stability to decrease in the series :

HC10 HC103 HC104

whereas it actually increases in the opposite direction.

Although the energy-content oi the molecule is the real factor
affecting stability, it is assumed that this internal energy is condi-
tioned by the mode of linkage of the atoms, i.e., by valency, and the
formulae of the above compounds are therefore usually written with
the chlorine atom possessing different valencies, from 1 to 7 :

/Cl /H

I. O< , or Cl— O— Cl ; O<f , or H— O— Cl.

XC1 XC1

in /O v^O

III and V. H— O— Cl = O, or H— O— Cl< | , or H— O— Clf

NO X)

iv^O O. v v^O
IV and V. Clf , or \C1— Cl/

°\vn vn,/0 /°\ vn/°

VII. O=C1— O— Cl =0 ; H— O—C1 O, or H— O— CJ= O

o/ \0 \o/ ^o

HOX vn/°

The hydrate HC1O 4,H,O may be written : ) Cl —OH

HO/ \Q

The variable valency of iodine, an element very similar to chlorine
and univalent in its stable compounds, is shown in the compounds
IC13, and KI3, in which iodine is assumed to be tervalent :

III /Cl in /K
Cl— I/ ; I— 1 .

392 INORGANIC CHEMISTRY CH. xxi

The alternative formulae for chloric and perchloric acids, given above,
are still undecided. The element manganese, which occurs in the same
group of the Periodic System as chlorine, is also assumed to be hepta-
valent in the compound potassium permanganate, KMnO4, which
resembles the perchlorate in crystalline form :

Vll/Q
KO— Mn=0.

EXERCISES ON CHAPTER XXI

1. What is the action of chlorine on (a) water, (b) a cold dilute solution
of caustic potash, (c) a concentrated solution of caustic potash,
(d) mercuric oxide, (e) dry slaked lime ? Give equations.

2. Starting with chlorine, caustic potash, and concentrated sulphuric
acid, how would you prepare (a) a solution of hypochlorous acid,
(b) chlorine monoxide, (c) perchloric acid, (d) chlorine dioxide ?

3. Describe the preparation and properties of the oxides of chlorine.
What constitutional formulae are attributed to them, and for what
reasons ?

4. How are chloric and perchloric acids prepared ? How would you
proceed to determine the formulae of these acids ?

5. Describe the manufacture of bleaching powder. What is the
formula of this material ? What is understood by the " Available
Chlorine " of bleaching powder, and how is it estimated ?

6. What happens when a solution of bleaching powder is (a) heated
alone, (b) saturated with chlorine and distilled, (c) heated with a little
cobalt chloride solution, (d) added to a solution of manganous chloride ?
What happens when solid bleaching powder is treated with (a) concen-
trated sulphuric acid, (6) hydrochloric acid ?

7. How are chlorates and perchlorates prepared on the large scale ?
For what purposes are they used ? By what reactions are chlorates
distinguished from perchlorates and from hypochlorites ?

8. What is the action of (a) concentrated hydrochloric acid, (b) con-
centrated sulphuric acid, (c) perchloric acid, on potassium chlorate ?
Describe the action of heat on this substance.

9. A mixture of 10 c.c. of chlorine, 10 c.c. of chlorine monoxide, 10
c.c. of chlorine dioxide, and 20 c.c. of carbon dioxide is treated with
caustic potash solution. What volume of gas remains ? The same
mixture is then heated, and the resulting gas treated with potash
solution. What are the volume and composition of the resulting gas ?

CHAPTER XXII

THE HALOGENS

BROMINE. Bn = 79'29.

Bromine. — Bromine was discovered by Balard (1826) in the
residues from the manufacture of solar salt (p. 220) at Montpellier.
These liquors are known as bittern, and contain magnesium bromide,
MgBr2. On the addition of chlorine, the liquid becomes yellow, and
gives an orange-red colour with starch-paste. Bromine is liberated
by displacement : MgBr2 -j- C12 = MgCl2 -|- Br2. If the bittern is
evaporated, and the residue distilled with manganese dioxide and
sulphuric acid, red vapours are evolved, condensing to a nearly
black liquid. This reaction suggests that the substance is similar to
chlorine. The name bromine (from Greek bromos, a bad smell) was
given to the substance on account of its unpleasant and powerful
odour. Bromine was at once recognised as a halogen, i.e., an element
of the same character as chlorine ; its discovery was further evidence
in favour of the elementary nature of the latter.

Bromide of silver, AgBr, occurs in certain Mexican and Chilean
silver. ores, but the source of the bromine of commerce is found in the
magnesium, sodium, potassium, and calcium bromides of certain
mineral springs. The Ohio springs contain 3-4—3-9 per cent, of
MgBr2. From these, the German springs of Kreuznach, Kissingen,
and Schonebeck, and the residues of the great potash deposits of
Stassfurt, practically all the bromine of commerce is made. Magne-
sium bromide occurs in sea- water, which contains 0-015 per cent, of
bromine ; the Dead Sea and the Great Salt Lake of Utah contain
considerable quantities of bromides. Bromides also occur in the
North wich brine. • Bromine is found in sea animals and plants ;
the ancient Tyrian purple, obtained from a shellfish, consists of
dibromindigo.

Preparation of bromine. — The most important bromine compound
in commerce is potassium bromine, KBr, used in photography, and in
medicine as a sedative. From this, bromine can be obtained by
heating with sulphuric acid and manganese dioxide (cf. chlorine) :
2KBr + Mn02 + 3H2SO4 = Br2 + 2KHS04 + MnSO4 + 2HaO.

393

394

INORGANIC CHEMISTRY

FIG. 199. — Preparation of Bromine.

EXPT. 160. — 2-5 gm. of powdered KBr, mixed with 7 gm. of MnO2, are
distilled in a retort with 15 c.c. of H2SO4 mixed 'with 90 c.c. of water.
The dark red vapour of bromine is condensed in a little water in the
receiver (Fig. 199). A red solution of bromine, bromine water, is

formed, and a
small quantity of
a nearly black
liquid settles out
at the bottom.
This is bromine.
The vapour acts
violently on the
mucous mem-
branes, so that
experiments with
bromine should be
carried out in a
good draught. It
also corrodes cork
and indiarubber.
The liquid should
be kept in a well-

stoppered bottle. It corrodes the skin, which should at once be
washed with petroleum if it comes in contact with bromine.

Bromine may be purified by careful distillation. Chlorine is
removed by distillation over potassium bromide : 2KBr -|- C12 ==
2KC1 -\- Br2. Iodine is removed as a precipitate of cuprous iodide,
Cul, by adding a solution of copper sulphate and sodium sulphite
to a solution of impure potassium bromide : 2CuS04 + Na2SO3 +
H2O = Cu2S04 (cuprous sulphate) -f Na2S04 '+ H2SO4 ; Cu2S04 +
2KI = 2CuI -f K2S04. Scott's method of preparation of pure
hydrobromic acid (p. 399) is the simplest way of obtaining a pure
bromine compound.

The manufacture of bromine. — A little bromine is made from
bittern by Balard's process. Chlorine is passed through, until
the yellow colour does not increase in intensity : MgBr2 -j- C12 =
MgCl2 -f- Br2. (Excess of chlorine is avoided, as it contaminates
the resulting bromine.) The bromine set free is shaken out with
paraffin oil, which dissolves it and floats to the surface. The oil is
then shaken with a solution of caustic soda, when sodium bromide
and bromate are produced, leaving the paraffin colourless and ready
for use over again : 3Br2 -f CNaOH = 5NaBr + NaBr03 +
3H2O. The aqueous layer is evaporated, heated to decompose the
bromate, and distilled with manganese dioxide and sulphuric acid.
Most of the bromine sold is prepared from residual liquors, con-

XXII

THE HALOGENS

395

taining magnesium bromide, from Stassfurt, or Ohio. These are
decomposed by chlorine in the apparatus shown in Fig. 200.

The liquor trickles down the tower, A , which is filled with earthen-
ware balls, and runs into a tank, B, provided with perforated shelves.
Steam is blown in at the bottom of this tank, and chlorine gas from
the generator, /), passes over the surface of the liquid, and up the
tower, meeting the descending liquid. The bromine is driven off by
the heat of the steam, and the vapour passes out of the top of the

tower to a cooling worm, where it
is condensed, the last traces of vapour

being kept back by moist iron filings
in a small tower, C. The bromide of
iron, Fe3Br8, so produced, is used
as a source of potassium bromide :
the solution is precipitated with

FIG. 200. — Manufacture of Bromine.

potassium carbonate : Fe3Br8+4K2C03+4H20=-8KBr+Fe3(OH)8
(black precipitate) -f- 4CO2.

Electrolytic methods have been used, but not to any great extent.

Properties of bromine. — Bromine is a dark red, almost black,
liquid, of high density (3-188 at 0° and 3-119 at 20°), which gives
off a dark red poisonous vapour, of most irritating odour. It
freezes to a dark red solid, melting at — 7-3° ; at — 252° this is
colourless ; the boiling point is 59°. The vapour density at 100° is
85 (H = 1) ; at 228° it is 79-6, corresponding with the formula Br2.
At lower temperatures there may be some polymerised molecules
present : Br4 ^ 2Br2. At 1050° dissociation into atoms to the extent

396

INORGANIC CHEMISTRY

CHAP.

of 6-3 per cent., and 30 per cent, at 1500°, occurs : Br2 ±=^2Br. Bromine
is a powerful irritant poison. It is used to some extent as a disin-
fectant, for which purpose it is absorbed in sticks of diatomite brick,
and the product (75 per cent. Br2) is called solid bromine. Bromine
is also used in synthetic organic chemistry, e.g., in the preparation of
eosin.

Bromine combines directly with many elements, forming bromides.

EXPT. 161. — Five c.c. of bromine are poured into a test -glass standing
inside a wide jar, open at both ends, over a draught -hole in the bench.
The top of the jar is closed by a glass plate having a small hole in the
centre (Fig. 201). A small piece of white phosphorus thrown into the
liquid causes an explosion, and is projected from the liquid. Red
phosphorus burns quietly with a lurid red flame, forming yellow fumes
of the pentabromide, PBr5. Powdered arsenic burns with a reddish-
white flame, forming fumes of AsBr3. A small piece of potassium com-
bines explosively, forming KBr Sodium,
however, does not combine with bromine
unless heated to 200° in the vapour, or
when water is added.

Bromine vapour bleaches moist
litmus paper, though more slowly than
chlorine. Starch-paste is coloured
orange-yellow by bromine water or
vapour. Bromine water is a solution
in water ; 3 '6 parts of bromine dissolve
in 100 of water at 20°; the solu-
bility decreases slowly with rise of
temperature. The red solution loses
bromine on exposure to air. The freezing point shows that the
bromine in solution has the formula Br2. Bromine water is stable
in the dark, but decomposes in bright sunlight : 2Br2 4- 2KLO =
4HBr + Oa.

If saturated bromine water is cooled in a freezing mixture, red
solid bromine hydrate, Br2,8H2O, separates. This decomposes at
17° into bromine water and bromine.

Chloroform, benzene, and carbon disulphide abstract bromine
from its aqueous solution, forming orange -red liquids.

EXPT. 162. — Add a little chlorine water to a solution of KBr, and shake
with chloroform. The latter separates out, containing most of the
bromine as a red solution. Shake this with caustic soda solution. The
chloroform becomes colourless, and the aqueous layer contains bromide
and bromate.

The atomic weight of bromine was found by Stas from the ratios
AgBrO3 : AgBr, and Ag : AgBr. Baxter (1906) synthesised AgBr,

FIG. 201.— Reactions with Bromine.

XXII

THE HALOGENS

397

and found Ag : AgBr = 0-574453 : 1. The conversion of AgCl into
AgBr gave AgBr : AgCl = 1-310171 : 1. These results agree with
Stas's values. The most accurate value has been found from the
density of hydrobromic acid, i.e., direct to H = 1, by Moles (1916).
The density of HBr at S.T.P. is 3-64442, and, after correction for
deviations from Boyle's law, this gives Br = 79-29.

Hydrobromic acid, HBr. — Bromine unites directly with hydrogen
when a mixture of the latter with bromine vapour is passed over

heated platinum : H,

Br2 = 2HBr.

The combination is not

attended with explosion, as in the case of hydrochloric acid, and does
not begin in the absence of a catalyst, even in bright sunlight, below
300°. In the presence of platinum, combination begins at 200°.
The heat of formation of HBr is only 12 kg. cal., as compared with
22 kgm. cal. with HC1, or 58 kg. cal. with H20.

EXPT. 163. — A current of dry hydrogen is passed through bromine in
a wash-bottle standing in water at 60°, and the mixed gas passed over a
heated spiral of
platinum wire in
a glass tube (Fig.
202). White fumes
are produced
when the gas
issues into moist
air. If passed into
water, as shown,
the gas is rapidly
absorbed, forming
a solution of hy-
drobromic acid.

FIG. 202.— Synthesis of Hydrobromic Acid.

Hydrobromic
acid is also de-
composed when

passed over heated platinum ; a state of equilibrium is set up :
2HBr ^ H2 -f Br2. An excess of hydrogen is used in the above
experiment, when combination is nearly complete.

Hydrobromic acid is most conveniently prepared by the action of
bromine on a mixture of red phosphorus and water. Phosphorus
tribromide and pentabromide are probably first formed, and at once
decomposed by water : PBr3 + 3H20 = H3PO3 (phosphorous acid)
-f 3HBr ; PBr5 -j- 4H20 = H3P04 (phosphoric acid) -f 5HBr.

EXPT. 164. — Twenty gm. of red phosphorus and 40 c.c. of water are
placed in a flask, and bromine is added drop by drop from a dropping
funnel (Fig. 203). The gas is passed through a U-tube loosely filled
with broken glass smeared with red phosphorus made into a paste with

INORGANIC CHEMISTRY

CHAP.

water. This removes unchanged bromine which volatilises. The
addition of the first few drops of bromine is attended by lambent green
flames, but when the air is displaced these disappear. The gas is
collected by downward displacement in dry gas jars. The jar is full
when dense fumes escape from the mouth, which is partially covered
with a glass plate. The gas may also be collected over mercury (cf.
hydriodic acid, p. 408). Commercial red phosphorus may contain arsenic,
and the HBr is then contaminated with AsBr3.

Hydrogen bromide may be obtained by the action of bromine on
benzene : C6H6 + 2Br2 = C6H4Br2 (dibromobenzene) -f 2HBr. This
is a reaction of substitution (p. 275) ; two atoms of hydrogen are
removed from the benzene molecule, and their place is taken by two
atoms of bromine. The two atoms of hydrogen form two molecules
of hydrogen bromide with two other atoms of bromine from the two

molecules of bromine
which took part in the
reaction. The molecules
of the halogen are there-
fore divided into two
parts ; one enters the
compound, and the
other combines with the
hydrogen atom which
is displaced.

Sixty-five c.c. of
bromine are dropped
slowly into 50 gm. of
dry benzene mixed with
a little aluminium pow-
der in a flask. The
reaction is started by
gentle warming, but when
evolution of gas com-
mences the flask is cooled. The gas is scrubbed in two U -tubes, the
first containing iron bromide, to remove Br2 vapour, and the second
anthracene, to remove benzene.

The physical properties of hydrogen bromide are as follows :
Melting point — 86° Density of liquid at b . -pt . 2 • 1 60

Boiling point — 68-7° Relative density of gas (H == 1) 40-1

Critical temperature + 91 -3° The three forms are colourless.
Normal density 3-644 gm. per lit.

Hydrogen bromide is very soluble in water ; 1 vol. of water dis-
solves 600 vols. of HBr at 10°. The solution is a strong acid :
^H' Br'.

FIG. 203.— Preparation of Hydrobromic Acid.

THE HALOGENS

399

EXPT. 165. — Collect a jar of the gas containing a glass bulb of water, and
fitted with stopcocks as shown in Fig. 204. Break the bulb by shaking.
Notice the fumes produced by the gas on contact with aqueous vapour.
Dip the vertical tube under water coloured with blue litmus, and open
the stopcock. The water rushes into the jar to fill the partial vacuum
created by the solution of the gas, and the litmus turns red. The
solution saturated at 0° contains 82, that at 15° 50, per cent, of HBr.

Aqueous hydrobromic acid may be prepared by passing the gas into
water. To prevent water being forced back into the generating
flask, on account of the great solubility of the gas, the latter may be
passed into the water through an inverted retort,
as shown in Fig. 205. If liquid is driven back, it
merely collects in the bulb of the retort.

Although concentrated sulphuric acid decom-
poses potassium bromide with the formation of
hydrobromic acid in the first instance, the gas soon
becomes mixed with bromine vapour, on account
of the oxidation of the hydrobromic acid by the
sulphuric acid : 2HBr + H2S04 = Br2 -f SO2 +
2H2O. If, however, 0-2 gm. of stannous chloride
and 3-4 c.c. of sulphuric acid are added to
25 c.c. of a solution of 15 gm. of' KBr and the
mixture distilled ; or if KBr is distilled with
syrupy phosphoric acid ; a solution of hydrobromic
acid, free from bromine, is obtained.

A solution of the acid is also obtained by-
passing sulphuretted hydrogen or sulphur di-
oxide through bromine covered with a layer of

water : Br,
2H20 ^ H

H2S = 2HBr

S; S02 + Br2 +
latter method

FIG. 204. — Experi-
ment to show
Solubility of Hy-
drobromic Acid in
Water.

04 + 2HBr. The
gives pure HBr (Scott, 1900).

Three hundred and fifty c.c. of bromine are covered
with 2 litres of water in a flask, and a current of SO2
from a siphon of liquid SO2 passed into the water
through a tube ending just above the surface of the bromine, until
the whole is converted into a pale yellow homogeneous liquid, which
is distilled. The liquid is redistilled over BaBr2 to remove sulphuric
acid carried over in the first distillation.

Concentrated hydrobromic acid fumes in moist air. On distilla-
tion it forms an acid of maximum boiling point, as in the case of
hydrochloric acid (p. 237). The composition of this liquid varies
from 47-38 to 47-86 per cent. HBr, according as the pressure during
distillation varies from 752 to 762 mm. ; it is not a definite hydrate.
The boiling point under 760 mm. is 126°.

400

INORGANIC CHEMISTRY

On cooling, two solid hydrates, HBr,2H2O, m.-pt. — 11-3°, and
HBr,4H2O, m.-pt. — 55-8°, are formed. A hydrate, HBr,H2O, has
also been described.

Aqueous hydrobromic acid is decomposed by oxygen in sunlight,
and becomes yellow from liberation of bromine : 4HBr -}- 02 —
2H2O -f 2Br2. A mixture of dry HBr and oxygen is not decom-
posed on exposure
to light. The gas
or solution is decom-
posed by chlorine .
2HBr -|- C12 =

2HC1 + Br2.
Bromides. — Hy-
drobromic acid is
extensively ionised
in solution, and is
almost as strong as
hydrochloric acid :
HBr =± H' + Br'.
It dissolves zinc,
iron, and many
other metals with
evolution of hydro-
gen, forming bro-
mides. The latter
may also be ob-
tained by neutral-
ising the acid with
oxides, hydroxides,
or carbonates, and
by the direct union
of the metals with
bromine. They are
ionised in solution :
KBr -^ K* + Br'.
The bromide ion, Br',
is contained in the solutions. Nearly all bromides are soluble in
water ; silver, lead, and mercurous bromides only very sparingly.
Silver nitrate solution is used as a test for bromides, i.e., for the
ion Br' : a yellowish-white precipitate of AgBr is formed, insoluble
in dilute nitric acid, and sparingly soluble in dilute ammonia (cf. AgCl
and Agl). Palladium nitrate gives a reddish-brown precipitate of
palladious bromide, PdBr2. The formation of free bromine, soluble
in chloroform with a red colour, by the action of chlorine water, and
the formation of red fumes of bromine when the substance is heated
with Mn02 and H2S04, are also characteristic reactions.

FIG. 205. — Preparation of Aqueous Hydrobromic Acid.

THE HALOGENS 40 1

Oxy-acids of bromine. — No oxides of bromine are known, but the
following oxy-acids have been described :

Hypobromous acid, HBrO.
Bromous acid, HBr02.
Bromic acid, HBr03. "

Perbromic acid, and its salts, are unknown.

Hypobromous acid, HBrO. — By shaking bromine water with
precipitated mercuric oxide, a solution of hypobromous acid, HBrO,
is formed. By adding more bromine, and mercuric oxide, a solution
containing 6 percent, of HBrO may be obtained : 2Br2 -f- 2HgO -f-
H2O = 2HBrO + HgBr2,HgO. The liquid may be distilled in a
vacuum at 40°. It is a yellow liquid, decomposing when heated:
4HBrO = 2H20 -f 2Br2 -f O2, and is a powerful oxidising and
bleaching agent.

If bromine is dissolved in cold aqueous potash or soda, unstable
hypobromites are formed : Br2 -{- 2NaOH = NaBr -f- NaBrO +
H20. These are used as oxidising agents and in the estimation of
hydrogen peroxide (p. 340) and of urea (p. 538). When the solutions
are kept, decomposition occurs, and a bromate is formed : 3NaOBr
= 2NaBr -f- NaBr03. Bromine vapour is also absorbed by dry
slaked lime, forming a red powder similar to bleaching powder.
This probably contains CaOBr2 ; when distilled with dilute nitric
acid, aqueous hypobromous acid passes over.

Bromous acid, HBr02. — This acid is said to be formed by the action of
excess of bromine water on a concentrated solution of silver nitrate.

Br2 + AgNO3 + H2O = HBrO + AgBr + HNO3.

2AgNO3 + HBrO + Br2 + H2O = HBrO2 + 2AgBr + 2HNO3.

Bromic acid, HBr03. — When bromine is dissolved in hot concen-
trated alkali a colourless solution is obtained which contains a
bromate and a bromide :

3Br2 + 6KOH = 5KBr + KBr03 -f 3H20.

Potassium bromate is much less soluble than the bromide, and the
two salts may be separated by crystallisation, as in the case of the
chlorate (p. 370). Potassium bromate also separates out when
bromine vapour is passed into a solution of potassium carbonate
which has been saturated with chlorine : 6KOC1 + Br2 = 2KBr03
+ 4HC1 + Cla.

If silver nitrate is added to a solution of potassium bromate,
silver bromate, AgBr03, is precipitated. If this is treated with
bromine water, insoluble silver bromide is formed, and the filtered
solution contains bromic acid : 5AgBrO3 -f 3Br2 -f 3H2O = 5AgBr
-f 6HBrO3. Bromic acid is also formed by passing chlorine through
bromine water : Br2 + 5C12 + 6H2O = 2HBr03 + 10HC1.

By evaporation on a water-bath, a 5 per cent, solution may be

D D

402 INORGANIC CHEMISTRY CHAP.

obtained. By distillation in a vacuum, a concentration of 50 per
cent, is reached. More concentrated solutions give off bromine and
oxygen : 4HBr03 = 2H2O -f 2Br2 -f- 5O2. Bromic acid is a power-
ful oxidising agent :

2HBr03 + 5S02 + 4H20 = Br2 + 5H2S04 ;
2HBr03 + 5H2S = Br2 + 6H2O + 5S ;
HBr03 + 5HBr - 3Br2 + 3H2O.

The bromates are usually sparingly soluble in water. On heating,
they decompose in one of two ways ; perbromates are not formed :

1. KBr03, HgBrO3, and AgBrO3, give bromide -f oxygen ;

2. Mg(Br03)2, Zn(Br03)2, Al(Br03)3, Pb(Br03)2 and Cu(Br03)2 give
oxide -f bromine -f- oxygen.

A mixture of NaBrO3 + SNaBr is prepared by saturating concen-
trated caustic soda with bromine, and draining the separated crystals.
To these sufficient NaBrO3, prepared by electrolytic oxidation of NaBr,
is added to form NaBrO3 -j- 2NaBr, and the mixture is used, under the
name of bromine salt, in the extraction of gold from its ores.

Barium bromate, Ba(BrO3)2, is precipitated when a slight excess
of bromine is added to hot concentrated baryta water : 6Ba(OH)2 -f
6Br2 = Ba(Br03)2 -f 5BaBr2 + 6H20. The bromide is soluble and
remains in solution. If barium bromate is digested with dilute
sulphuric acid, and the excess of the latter removed by baryta water,
the filtered solution contains bromic acid.
•

IODINE. I = 125-91.

Iodine. — In 1812 Courtois, of Paris, discovered that the mother-
liquors from which soda had been crystallised in the manufacture
from kelp, or seaweed-ashes, gave off a violet vapour when heated
with manganese dioxide and sulphuric acid. These vapours con-
densed to a black metallic-looking crystalline substance. The in-
vestigation of this material, called " the substance X," was begun
by Gay-Lussac and simultaneously by Davy, who, by permission of
Napoleon, was passing through Paris to Italy at the time. Davy
published his results on December llth, 1813, and Gay-Lussac a
day later. The substance was recognised by these investigators
as a new element analogous to chlorine, and received the name
iodine (from the Greek ia'ides, violet-coloured) on account of the
beautiful violet colour of its vapour (p. 10). They showed that it
formed a hydrogen compound, hydriodic acid, HI, exactly analogous
to hydrochloric acid.

Iodine, like chlorine and bromine, occurs only in combination.
(Free iodine is said to exist in the water of Woodhall Spa, near
Lincoln, North America.) Its compounds with metals, called

xxn THE HALOGENS 403

iodides, occur, in small amounts but widely diffused, in the three
kingdoms of Nature. The iodine content of sea- water, which exists
partly as organic compounds and partly as iodides, is small. It
never exceeds 0-001 per cent., and in the Atlantic is only 1 part in
280 millions. Seaweeds and sponges absorb this iodine in the form
of organic compounds (e.g., iodospongin) : tropical sponges may
contain as much as 10 per cent, of iodine, whilst Turkey sponges
contain about 0-2 per cent. The amount of iodine is greater in
deep-sea weeds than in those growing near the shore. During storms,
these weeds are torn up and cast ashore. They are known in Scot-
land as drift-weeds, or red wracks ; the varieties known as Laminaria
digitata and L. stenophylla alone are used in the manufacture of
iodine.

The weeds are burnt in shallow pits, and the ashes, known as
kelp (varec in Normandy), contain potassium salts and from 0-4 to
1 -3 per cent, of iodine as iodides. Formerly, in Normandy, Spain, and*
Scotland, these ashes were used in the manufacture of alkali (potash) ;
the manufacture of iodine was begun by Dr. Ure at Glasgow, and
three works are still in operation in that town. Iodine manufacture
is also carried out from seaweed in Norway and Japan.

Iodine occurs in oysters and many sea-animals. It is present in
traces in cod-liver oil, as an organic compound, and occurs as an
organic compound iodothyrin. CijH^OgNIg, in the thyroid glands
(especially of the ray and dogfish, which contain 1 per cent, of iodine).
In the mineral kingdom iodine occurs in certain lead and silver ores,
and in some magnesian limestones and dolomites. The deposits of
seaweed in strata in Central Europe contain iodine, and the water
which percolates to them appears in springs which contain iodides,
such as those of Heilbrunn, and of Montpellier, which are used
medicinally.

The iodine of the body seems to be absorbed in the lungs from the
spores of lower organisms floating in the air ; normally about
0-005 mgm. of I passes into the lungs in this way per twenty-four
hours.

The most important source of iodine is the sodium iodate contained,
to the extent of 0-2 per cent., in crude Chile nitre (caliche). The
mother-liquors from the crystallisation of the nitrate contain about
3 gm. of iodine as iodate per litre.

Preparation of iodine. — In the laboratory, iodine may be obtained
by heating potassium iodide with sulphuric acid and manganese
dioxide : 2KI + Mn02+ 3H2S04 = I2+ 2KHS04 + MnSO4+ 2H20.

EXPT. 166.— Heat 3-5 gm. of KI with 7 gm. of MnO2 and 100 c.c.
of dilute H2SO4 (1:6) in a retort. Beautiful violet vapours are given
off, which condense in the neck of the retort and in the receiver as glitter-
ing black scales of solid iodine.

D D 2

404

INORGANIC CHEMISTRY

CHAP.

In the manufacture of iodine the kelp is lixiviated with water
in iron vats, heated by steam, and the solution concentrated in
iron pans. The salts which separate, called plate sulphate, con-
sisting chiefly of potassium sulphate, are fished out. On cooling,
impure potassium chloride separates, and on further evaporation
crude sodium chloride (" kelp salt ") is deposited. The final mother-
liquor contains the very soluble sodium and potassium iodides,
together with some bromides. It is mixed with sulphuric acid, and
the sulphur, from the decomposition of sulphides, allowed to settle.
The clear liquor is then run into the iodine stills, consisting of iron

FIG. 206. — Manufacture of Iodine.

pots with dome-shaped lead covers communicating with trains of
earthenware receivers, called udells (Fig. 206). Manganese dioxide
is added, and iodine distils off on heating, collecting in the udells.
It is purified by sublimation in porcelain pans. About 12 Ib. of
iodine are obtained per ton of kelp, representing about half that
contained in the original weed.

The two processes of Stanford (1863), established in the Outer
Hebrides, are no longer worked. In the char process, the sun-dried
weed was distilled in iron retorts at a low red heat. It was expected
that acetic acid and tar would be recovered, but only a little evil-
smelling tarry water came over. The residue was lixiviated. In
the wet process, the weed was boiled with sodium carbonate solution,

xxn THE HALOGENS 405

and filtered. Fairly pure cellulose, amounting to 15 per cent, of
the weed, was left. This was called algulose, and was used for
making paper. On acidifying the filtrate, a gelatinous substance
called algin was thrown down, which was used in making jellies,
sizing paper, and as a glue. The filtrate, containing iodides, was
evaporated, neutralised with limestone, and distilled with sulphuric
acid and manganese dioxide. This process seems to have been
recently revived in Norway ; the algin is sold as Norgine for use as
an adhesive.

In France, the kelp liquors are acidified with hydrochloric acid, and
chlorine is passed in. Iodine is precipitated ; it is filtered off, washed,
dried, and resublimed in earthenware retorts : 2KI + C12 = 2KC1 + I2.

The main source of iodine at the present day is the mother-
liquor (" aqua vieja ") of caliche. This contains about 4-5 gm. of
sodium iodate, NaI03, per litre. It is run into lead-lined vats, and
treated with dilute sulphuric acid and sodium hydrogen sulphite,
lodic acid is first liberated, and is then reduced by sulphurous acid : *

(1) NaI03 -f H2S04 = NaHS04 + HI03.

(2) 2HI03 + 5H2S03 = I2 + 5H2S04 -f H20.

The liberated iodine at first reacts with the excess of sul-
phurous acid, and it is only at the end of the reaction, when the latter
is used up, that iodine appears :

(3) I2 + H20 -f H2S03 = 2HI -f H2SO4.

(4) HI03 + 5HI = 3H20 + 3I2.

The iodine precipitated is pressed, washed, and resublimed.

The above process involves the mutual decomposition of iodic
acid and hydriodic acid : the former is an oxidising agent and the
latter a reducing agent. This reaction sets in only when all the
free sulphurous acid is used up, and the whole process therefore
exhibits a period of induction (p. 235). This is very clearly shown in
the following experiment, due to Landolt.

EXPT. 167. — Dissolve 10 gm. of crystallised iodic acid in 1 litre of
water. Saturate 5 c.c. of water with sulphur dioxide, and add the
solution to 1 litre of water. 50 c.c. of the iodic acid solution are added
to 250 c.c. of water in a cylinder, and a little starch solution is added.
50 c.c. of the sulphurous acid are diluted with 250 c.c. of water in a
cylinder, and the solution is poured quickly into the iodic acid. The liquid
remains colourless for a certain interval, and then at once becomes blue.
By varying the dilution, the time interval may be altered. This is
an example of successive reactions; the later reactions use up
the products of the first, and the speed of the whole reaction is that of
the slowest component reaction.

406 INORGANIC CHEMISTRY CHAP.

Pure iodine. — Commercial iodine nearly always contains iodine
chloride, IC1, iodine bromide, IBr, and sometimes cyanogen iodide,
ICN. All these substances are volatile, and cannot be separated
by sublimation. Resublimation over potassium iodide removes
most of the impurity.

EXPT. 168. — A little iodine is ground up in a mortar with potassium
iodide, and the mixture gently heated in a porcelain dish on a sand-bath
A larger porcelain dish, filled with cold water, is placed over the first
one, and the purified iodine condenses on its under surface in glittering
scales with a brilliant metallic lustre.

Stas dissolved resublimed iodine in a strong solution of KI, precipitated
it with water, and distilled it in steam The solid iodine which came
over was collected, dried in vacuo over solid calcium nitrate (frequently
renewed), and finally sublimed over caustic baryta, BaO, to separate
HI and H2O. Ladenburg (1902) washed precipitated silver iodide with
dilute ammonia to free it from chloride, reduced it with zinc and dilute
sulphuric acid, Agl + H = Ag -f- HI, precipitated the iodine from the
solution with nitrous acid : 2HI -f 2HNO2 = 2H2O + 2NO -f I2,

distilled it in steam, and dried
it over calcium chloride. Lean
and Whatmough (1900) heated
pure cuprous iodide to 240° in a
current of dry air : Cu2I2 + O2 =
2CuO + I2.

Properties of iodine. — Iodine
is a blackish-grey crystalline
solid which is opaque, and
has almost a metallic lustre.

FIG. 207.-crystais of iodine. (When deposited in thin films

on glass at— 180° it is trans-
parent.) Large crystals, belonging to the rhombic system (Fig. 207),
are produced by the spontaneous evaporation of the ethereal
solution, or by allowing hydriodic acid to oxidise by exposure to
air. The physical properties of iodine have been differently stated :

Stas : — Ladenburg : —

Sp. gr. 4-948 (17°). 4-933 (4°/4°)

Melting point 114-2° (solidif. at 113-6°). 116-1°

Boiling point 184*35° (Ramsay and Young). 183-05°.

Iodine vapour when pure has a splendid deep-blue colour ; when
mixed with air it is reddish -violet (Stas).

The density of iodine vapour diminishes with rise of temperature.
At the boiling point it corresponds with the formula I2 ; this remains
practically constant up to 700°, but then diminishes up to 1700°,
when it again becomes constant and corresponds with the formula I.
The dissociation into atoms : I2 ^ 21, which is doubtful in the

xxn THE HALOGENS 407

case of chlorine and bromine, is therefore well established with
iodine.

Iodine vapour shows an orange-yellow fluorescence, especially when
exposed to green rays. When exposed to the light from a mercury
lamp, it emits a complicated resonance spectrum, consisting of a large
number of equally -spaced lines.

Iodine is much less soluble in water than either chlorine or bromine ;
1 part dissolves in 3616 of water at 18°, 2145 parts at 35°, and 1084
parts at 55°. The solution has a brownish -yellow colour, and appears
to undergo slight decomposition on standing : 2I2 -f 2H2O —
4HI -f 02. For this reason the solubility of iodine slowly increases,
since the element is readily soluble in solutions of hydriodic acid or
iodides, forming dark brown liquids containing the ion I3'. From
the solution in potassium iodide, black crystals of potassium tri-
iodide, KI3, may be obtained. Chloroform and carbon disulphide,
which readily extract iodine from aqueous solutions, forming purple
solutions, do not do so from solutions in potassium iodide. The
compounds CsI3, CsI5, RbT3, and KI7 are known.

Iodine is readily soluble in alcohol, forming a brown solution known
as tincture of iodine (J oz. of iodine -f- i oz. of potassium iodide -(-
1 pint of rectified spirit). The solution in ether is also brown, and
it is suggested that in these solutions the iodine is in combination
with the solvent. The depression of freezing point of methylene
iodide, CH2I2, containing dissolved iodine, gives the formula I2.

Iodine combines directly with many elements, such as phos-
phorus (p. 18), and mercury (p. 116), forming iodides.

Test for iodine. — Solutions of iodine give a beautiful blue colour
with starch-paste. The latter is prepared by boiling " soluble

trch " with water, or adding boiling water to ordinary starch
lade into a paste with cold water. 1 part of iodine in 450,000 parts

water may be detected by this reaction. The blue colour dis-
ippears on heating, but reappears on cooling.

EXPT. 169. — Add a drop of iodine solution to some starch solution in
test-tube. Dip the lower part of the. tube containing the blue liquid
into a beaker of boiling water : the lower part becomes colourless. Cool
under the tap ; the whole again becomes blue. If excess of chlorine
water is added, the blue colour again disappears, since iodine chloride,
[Cl, is formed.

The blue substance has been variously supposed to be a chemical
>mpound— " iodide of starch " — or a solid solution, or an adsorption
>mplex of starch and iodine. A blue colour is produced by the
Jtion of iodine on other substances, e.g., saponarin, some of which
crystalline, and it appears only in the presence of iodides, or
lectrolytes. Basic lanthanum and praseodymium acetates, which
colloidal, also give a blue colour with iodine.

408

INORGANIC CHEMISTRY

CHAP.

Hydriodie acid, HI. — Hydrogen and iodine combine only with a
very feeble affinity. The affinity for hydrogen diminishes very
rapidly in the series : Cl, Br, I. A mixture of iodine vapour and hy-
drogen passed over heated spongy platinum forms hydrogen iodide,
HI, giving fumes in moist air, but the reaction is reversible and in-
complete : H2 -f I2 ^ 2HI.

Hydriodie acid may be obtained by heating potassium iodide with
phosphoric acid ; with sulphuric acid oxidation occurs, iodine being
set free, and the sulphuric acid is reduced to sulphuretted
hydrogen (cf. HBr) : H2S04 + SHI == H2S + 4H20 -f 4I2. The usual
method of preparation is by the action of water on a mixture of red
phosphorus and iodine : 2P + 5I2 + 8H2O = 10HI + 2H3PO4.

Phosphorus iodides are probably
first formed, and then decomposed
by water, as in the preparation of
hydrobromic acid.

EXPT. 170. — Four gm. of red
phosphorus and 20 gm. of iodine
are shaken together in a flask, and
water is slowly dropped on the
mixture from a tap -funnel (about
15 c.c.). The evolution of gas
may become very rapid, and the
flask is then cooled. The .gas is
collected directly by displacement
(Fig. 208). It is very soluble in
water, and attacks mercury.

Hydrogen iodide is a colourless
gas, very soluble in water (425
vols. HI in 1 vol. at 10°), and
fuming strongly in moist air. The
solubility may be demonstrated
by EXPT. 165. The gas condenses to a liquid under 4 atm. pressure at
0°, and is therefore much more easily liquefied than HC1 or HBr.
The physical properties of HI are as follows :

FIG. 208. — Preparation of Hydrogen
Iodide.

Boiling point —35-5°
Melting point —50-9°

Relative density (H = 1) 63-94
(theoretical for HI = 63-45)

The volumetric composition of the gas, as well as that of hydrogen
bromide, may be demonstrated by the action of sodium amalgam, as
in the case of hydrogen chloride. Half the volume of hydrogen
remains.

A jet of hydrogen iodide may be burnt in oxygen, with liberation
of violet fumes of iodine : 2HI = H2 -f I2 ; 2H2 + 02 = 2H2O.

Hydrogen iodide is decomposed by exposure to sunlight : after

xxn THE HALOGENS 409

ten days Victor Meyerjbund 60 per cent, decomposed ; after a
year, 99 per cent. : HI -- H -{- I- The decomposition is also readily
brought about by heat : 2HI ^ H2 -f- I2 : a hot glass rod placed
in a jar of the gas liberates violet fumes of iodine. The decom-
position begins at 180°, but is then very slow. The rate of decom-
position is quicker the higher the temperature. At each tem-
perature a fixed amount of decomposition is ultimately reached,
and then remains constant, i.e., a state of equilibrium is attained :
2HI=:±H2 + I2. At 350°, 17-3 per cent., at 444°, 79 per cent.,
of the gas is decomposed. The limit of decomposition at 250° is
reached only after several months, but at 444° it is attained after
a few hours. The rate of reaction is considerably accelerated by
the presence of spongy platinum, which acts as a catalyst. The
reverse reaction : H2 -f I2 ^ 2HI, also proceeds slowly, but is
accelerated by platinum. At 444°, the reaction stops when 21 per
cent, of hydriodic acid is formed, and therefore 79 per cent, of the
hydrogen and iodine vapour (in equal volumes) remains uncombined.
Thus, the same equilibrium state is attained, at a given temperature,
from the mixture of hydrogen and iodine vapour (H2 -{- I2) as from
hydriodic acid (2HI). This is characteristic of truly reversible
reactions. The catalyst produces no change in the composition
of the equilibrium mixture, since it accelerates equally both the
direct and inverse reactions.

Aqueous hydriodic acid is produced by dissolving the gas in
water. The apparatus shown in Fig. 205 may be used to prevent
the liquid being drawn back into the flask, owing to the great
solubility of the gas.

The solution saturated at 0° has a sp. gr. of 1-99, and contains
90 per cent, of HI. The hydrate HI,2H2O, m.-pt. — 43°, separates
on cooling. The solution ordinarily used in organic chemistry as a
reducing agent has a sp. gr. of 1-5.

An acid of maximum boiling point 127° at 76 cm. contains 57
per cent, of HI. The aqueous solution when freshly prepared is
colourless, but rapidly becomes brown when exposed to air, owing
to formation of iodine, which dissolves in the acid :

4HI + 02 = 2H20 + I2.

The ease with which this reaction occurs renders the concentrated
aqueous acid a valuable reducing agent.

Chlorine, or bromine, water readily liberates iodine from the acid :

2HI + C12 = 2HC1 + I2 ; or 21' -f C12 = 2C1' + I2.

Aqueous hydriodic acid is also formed by passing sulphuretted
hydrogen through a suspension of iodine in water :

H2S + I2 = 2HI + S ;
the sulphur is filtered off. When the density of the solution reaches

410 INORGANIC CHEMISTRY CHAP.

1-56, the action ceases. Sulphuretted hydrogen gas does not act
upon dry iodine, but the heat of solution of the hydrogen iodide
in water supplies the energy necessary for the reaction :

H2S = H2 + S - 2 -7 kg. cal.
H2 + I2 (solid) == 2HI 12 kg. cal.

H2S + I2 2HI + S - 19-7 kg. cal.

The heat of solution of 2 HI in a large quantity of water is 384
kg. cal., hence the heat evolved at the beginning of the reaction is :

H2S + I2 (solid) + Aq. = 2HI,Aq. -f S + (384 - 19-7) kg. cal.

As the solution becomes concentrated, the heat of solution of the
hydrogen iodide becomes less, until at sp. gr. 1-56 it is only 19-7
kg. cal. for 2HI. Further action then ceases.

Although the heat of formation of hydrogen iodide from hydrogen
and solid iodine is attended with an absorption of heat, the reaction
H24- I2 (vap.) = 2HI is attended with a slight evolution of heat, i.e., a
little heat is absorbed when hydrogen iodide dissociates into hydrogen
and iodine vapour. The extent of dissociation therefore increases with
the temperature (p. 355).

Chlorides of iodine. — Iodine monochloride is formed by passing
chlorine over iodine : I2 -f- C12 = 2IC1. A dark red liquid is formed,
which solidifies on standing, especially in contact with a trace
of IC13. The first product of solidification melts at 14°, but is
unstable, and is converted on standing into another stable modifi-
cation, melting at 27 '2°, which forms beautiful red needles. This
is the stable form under all conditions ; from the liquid cooled below
14°, crystals of either form separate according as a crystal of one
or the other form is added. The unstable form is obtained by
cooling the liquid to — 10°.

Iodine monochloride is decomposed by water :

5IC1 -f 3H20 = 5HC1 + 2I2 + HI03 (iodic acid) ;

alkalies decompose it into chloride, iodate, and iodide. It is also
formed by dissolving iodine in aqua regia, and extracting with
ether, or by heating iodine with potassium chlorate. It boils at
101 -3°, and the vapour density is normal.

Iodine trichloride, IC13, is obtained by the action of excess of
chlorine on iodine, or on the monochloride : IC1 + C12 — IC13.
The latter reaction is reversible, since the vapour density of the
trichloride shows that it is dissociated ; the decomposition is com-
plete at 670°. It may be volatilised in an atmosphere of chlorine.
The trichloride is also produced by heating iodine pentoxide in

xxn THE HALOGENS 411

hydrogen chloride : I2O5 -f 10HC1 = 2IC18 + 5H20 + 2C12. It is
a lemon-yellow crystalline solid, which is decomposed by alkalies
in the same way as the monochloride.

EXPT. 171. — If a jar of hydrogen iodide is inverted over a similar
jar of chlorine, and the glass plates are withdrawn, there is a violent
reaction, and dense fumes are formed. On standing, three substances
are seen to have been formed : (i) a violet vapour, depositing solid
crystals of iodine in the upper jar ; (ii) dark red drops of liquid at the
junction of the two jars — this is iodine monochloride, IC1 ; (iii) lemon-
yellow crystals in the lower jar — these are iodine trichloride, IC13.
The reaction is :

4HI + 4C12 = 4HC1 + I2 + IC1 + IC18.
On standing, only yellow crystals of IC13 remain.

Iodine trichloride may be regarded as a salt ; iodine acetate,
I(C2H3O2)3, is obtained by the action of C12O on iodine dissolved
in glacial acetic acid, and a sulphate, I2(SO4)3, and perchlorate,
I(C1O4)3,2H2O, have been prepared. The latter is obtained in yellowish-
green needles by the action of ozone on a cooled solution of iodine in
anhydrous perchloric acid : I2 + 6HC1O4 + O3 = 2I(C1O4)3 + 3H2O.
The strongly basic diphenyl-iodonium hydroxide, (C6H5)^IOH, is stable,
and forms salts which resemble those of tervalent thallium (p. 904),
even to giving a green flame reaction.

A stable pentafluoride, IF6, m.-pt. — 8°, b.-pt. 97°, is formed directly
from the elements, and is of interest in demonstrating the quinque-
valence of iodine in some of its compounds (cf. iodic acid).

Oxides and oxy-acids of iodine. — The following oxy-compounds
of iodine are known :

Oxides. Oxy-acids.

Hypoiodous acid, HOI
Iodine dioxide, I02 or I204
Iodine pentoxide, I205 Iodic acid, HIO3

Periodic acid, HI04,2H20 orH5IO6

A number of salts of periodic acids of different formulae are known.

Oxides of iodine. — The best-known oxide of iodine is the pent-
oxide, but two lower oxides, I4O9 and I02 have been described.

A green oxide, I4O9 (p. 329), is said to be formed by the action of
ozone on dry iodine. The dioxide, IO2, or I2O4, is obtained as a
lemon-yellow powder by the action of cold nitric acid on iodine, or by
the action of hot concentrated sulphuric acid on iodic acid. It de-
composes into its elements at 130°.

Iodine pentoxide, or iodic anhydride, I205, is obtained by heating

412 INORGANIC CHEMISTRY CHAP.

iodic acid to 200° : 2HI03 = H20 + I2O5. It forms white scaly
crystals, decomposing at 300° after fusion, into oxygen and iodine.
It oxidises carbon monoxide on warming, even if this gas is con-
tained only in traces in gaseous mixtures : 5CO + ^2^5 = 5C02+I2.
The carbon dioxide formed may be absorbed by baryta water, and
the amount determined by titration. Iodine pentoxide dissolves
in water, forming iodic acid, HI03. It is the anhydride of this acid.
Hypoiodous acid, HOI. — Iodine dissolves in cold dilute alkali
to form a yellow solution, with a characteristic odour of saffron.
This contains free hypoiodous acid, HOI. Probably a hypoiodite
is first formed, but this is almost completely hydrolysed by water,
even in presence of excess of alkali :

I2 + 2KOH ^ KI + KOI + H2O.
KOI + H20 ^± HOI + KOH.

The reaction involves the hydrolysis of the iodine molecule :
I2 + H2O — HI -f HOI. The compound HOI appears from this
equation, and from its properties, to be a feeble base rather than an
acid. The existence of a lower oxide of iodine in the freshly-
prepared solution of iodine in alkali may be inferred from the colour
and smell of the solution, and its strong oxidising and bleaching
properties. Indigo solution is bleached, hydrogen peroxide evolves
oxygen, manganous sulphate is precipitated as brown manganic
hydroxide, Mn(OH)3, and if alcohol is added to the solution a yellow
precipitate of iodoform, CHI3, is formed :

C2H5-OH + 4I2 + 6KOH =

CHI3 + HCO2K (potassium formate) -f 5KI + 5H20.
On standing, especially if heated, the alkaline solution of iodine
loses all these properties, and contains only an iodide and iodate :
3KOI = KI03 -f- 2KI. Free hypoiodous acid is also formed on
shaking an aqueous solution of iodine with precipitated mercuric
oxide : 2HgO + 2I2 + H2O = HgI2,HgO + 2HOI.

Iodic acid, HI03. — This, the most important oxy-acid of iodine,
is formed by the oxidation of the latter with ozone in presence of
water, or by heating iodine with ten times its weight of nitric ucid
(sp. gr. 1-5) in a flask on a water-bath, evaporating to dryness,
heating to 200° to expel nitric acid, and dissolving the iodine
pentoxide formed in the smallest amount of warm water. On
cooling the syrupy liquid, colourless rhombic crystals of iodic acid
separate.

It is also formed by passing chlorine through a suspension of
iodine in water : I2 + 5C12 + 6H2O = 2HI03 + 10HC1. Hydro-
chloric acid is removed by evaporation, or by addition of silver
oxide, when insoluble silver chloride is formed.

Iodic acid is insoluble in alcohol, but is very soluble in water,
and is deliquescent. The solution first reddens, and then bleaches

xxn THE HALOGENS 413

litmus paper. The solid deflagrates when heated with powdered
charcoal, sulphur, phosphorus, or organic matter. It is an oxidising
agent :

2HI03 + 5S02 + 4H2O = I2 + 5H2SO4 ;

2HIO3 + 5H2S = I2 + 6H2O -f 5S ;

HI03 + 5HI = 3I2 + 3H2O.

If iodine is dissolved in aqueous caustic potash, an iodate is formed :
3I2 + 6KOH = 5KI + KIO3 + 3H2O (Davy, 1815). If an acid is
now added, the whole of the iodine is again set free, on account
of the reduction of the iodic acid by the hydriodic acid. If neutral
solutions of iodate and iodide are mixed, an acid may be estimated
by adding it to this solution, and titrating the iodine liberated.

If iodine is added to a hot concentrated solution of potash,
potassium iodate, KI03, crystallises out on cooling, as it is sparingly
soluble. The salt may also be prepared by heating iodine with
potassium chlorate, or by adding iodine to a hot concentrated
solution of potassium chlorate, and boiling with a few drops of nitric
acid. Chlorine is expelled : 2KC103 + I2 = 2KI03 + C12. If
barium chloride is added to a solution of potassium iodate, barium
iodate is precipitated. This is decomposed by dilute sulphuric
acid, forming iodic acid : Ba(I03)2 + H2S04 = BaSO4 + 2HI03.
Iodic acid forms three series of salts, viz., normal salts and two
acid salts :

Normal potassium iodate, KI03 ;

Acid potassium iodate, KI03,HIO3, or KHI2O6 ;

Diacid potassium iodate, KI03,2HI03.

The acid salts are isomorphous with acid salts of some dibasic
organic acids (succinic, etc.), so that the acid is sometimes regarded
as dibasic, H2I206. The normal iodates are insoluble, or sparingly
soluble, in water. On heating, they break up hi one of two ways :
(i) into iodide -f- oxygen ; (ii) into oxide -f iodine -j- oxygen-
Barium iodate forms a periodate (see below). Iodates form complex
compounds with molybdic, tungstic, and phosphoric acids, and
with selenates.

Iodates are detected by the blue colour, due to liberation of
iodine, produced when sulphur dioxide is passed through the
solution, to which starch-paste has been added.

v>° .

The formula of iodic acid is assumed to be HO — ~L? , in which

iodine is quinquevalent. This does not, however, explain the
formation of the acid salts.

Periodic acid, HI04,2H20, or H5I06. — If a concentrated solution
of iodic acid is electrolysed at low temperatures, with a lead plate
covered with lead peroxide as anode, enclosed in a porous cell,

414 INORGANIC CHEMISTRY CHAP.

and a platinum plate immersed in dilute sulphuric acid as a cathode
outside, it is oxidised to periodic acid, HI04. The solution yields
colourless, deliquescent crystals, of the formula HI04,2H20. The
anhydrous acid is not known, and the water cannot be removed
from the acid without decomposing the latter, so that the formula
is probably H5IO6, salts of which are known. The crystals melt
at 133°, and decompose at 140° : 2H5IO6 = I2O5 + 5H20 -j- O2.

The solution is strongly acid, and is an oxidising agent, but it
does not oxidise sulphur dioxide.

If a solution of potassium iodate, to which a little potassium
chrbmate has been added, is electrolysed as described, sparingly
soluble potassium periodate, KIO4, is formed. An acid sodium
periodate, Na2H3I06, is formed by oxidising a boiling solution of
13 gm. of iodine, in a 10 per cent, solution of 60 gm. of caustic
soda, with a rapid stream of chlorine. The salt is precipitated.
A solution of this salt gives with silver nitrate a black precipitate of
the silver salt, Ag3IO5, which is decomposed by chlorine in presence
of water, giving silver chloride and a solution of periodic acid.

Barium periodate, Ba5(I06)2, is very stable, and is formed on heating
barium iodate to redness :

5Ba(I03)2 - Ba5(I06)2 + 4I2 + 9O2.

It is decomposed by dilute sulphuric acid, with formation of periodic
acid.

The periodates appear to be derived from acids formed by the
addition of water to a hypothetical anhydride, I2O7, in which iodine
is septavalent :

I2O7 + H2O = 2HIO4, forming meta-periodates, e.g., KIO4, AgIO4 ;
I2O7 -f 2H2O = H4I2O9, forming diperiodates, e.g., Na4I2O9 ;
I2O7 + 3H2O = 2H3IO5, forming mesoperiodates, e.g., Ag3IO5 ;
I2O7 + 5H2O = 2H5IO6, free paraperiodic acid, forming, e.g.,
Ba5(IOc)2.

The atomic weight of iodine. — The atomic weight of iodine is of
considerable theoretical importance in connection with that of
tellurium (p. 533), and several exact determinations of its value
have been made. Stas determined the ratio Agl : O by decom-
posing silver iodate by heat, and absorbing the oxygen in red-hot
copper. He also determined the ratio Ag : Agl = 100 : 217-534,
which, with the value 107-94 (O = 16) for silver, gave I = 126-86
(O = 16). This value, agreeing exactly with that of Marignac,
determined by the same method, was later found to be about half
a part per thousand too small, by reason of the occlusion of silver
nitrate in the precipitated silver iodide. More recent determina-
tions, all based on the value for silver, give I = 126-92 (O = 16),
or 125-91 (H = I).

XXII

THE HALOGENS

415

FLUORINE. F == 18-9.

Fluorine. — The mineral fluorite, or fluorspar, occurs in Derby-
shire, crystallised in cubes or octahedra (Fig. 209), or in compact
masses, like marble. It is known as " Derbyshire Spar," or, when
coloured blue or purple, as " Blue John." Colourless, transparent
crystals exhibit a bluish tinge when light falls on them, and this
property, which is shown by petroleum, solutions of quinine salts,
and other substances, is therefore known as fluorescence (cf. p. 8).
Fluorspar occurs in many other localities, and has long been used
in metallurgy as a flux, i.e., a substance which forms with the earthy
portions of ores (gangue) a readily fusible slag, which flows away
from the
metallic part
on reduction,
leaving the
metal as a
separate
layer. For
this reason
the mineral
obtained its
name, from
the Latin
fluo, I flow.

The com-
position of
fluorspar was
for a long
time un-
known. About 1670, Schwankhardt, of Nuremberg, observed that
a mixture of powdered fluorspar and concentrated sulphuric acid
corroded glass, and in 1771 Scheele discovered that the mineral was
a salt of lime and a peculiar acid, which he obtained in an impure
state by distilling fluorspar with concentrated sulphuric acid in a tin
retort. A glass retort was powerfully corroded, and a gas formed
which deposited gelatinous silica on passing into water. Gay-Lussac
and Thenard investigated the acid in 1809 ; they regarded it as the
oxide of an unknown radical. Ampere, in 1810, suggested that it
was probably a compound of hydrogen with an unknown element,
fluorine, analogous to hydrochloric acid. Fluorspar would then be
calcium fluoride, CaF2. The element was first isolated by Moissan
in 1886.

Fluorine is widely distributed in Nature, both in the form of
fluorspar and in other fluorides. Large quantities of cryolite, a
double fluoride of sodium and aluminium, AlF3,3NaF, are found in

FIG. 209.— Crystals of Fluorspar.

416

INORGANIC CHEMISTRY

CHAP.

Greenland, and fluor-apatite, CaF2,3C'a3(PO4)2, is a common mineral.
Small quantities of calcium fluoride in the soil, probably derived
from apatite, are absorbed by plants, the ashes of which contain
about Ol per cent, of fluorine. From plants, calcium fluoride
passes into the bones and teeth of animals, especially into the hard
parts ; the enamel of teeth may contain 0-3 per cent, of fluorine,
possibly in combination as apatite.

Traces of free fluorine seem to occur in varieties of fluorspar, such
as that of Wolsendorf, which have been decomposed by the radio-
active uranium minerals found in the same localities. The blue colour
of some kinds of fluorspar is apparently due to organic matter. On

FIG. 210. — Moissan's Apparatus for Preparing Fluorine.

heating, the blue colour disappears. Colourless fluorspar also becomes
blue when exposed to radium emanation.

The isolation of fluorine. — The isolation of fluorine was for a long
time one of the master problems of inorganic chemistry. The
attempts of Davy, Fremy, Nickles, Louyet, and Gore were uni-
formly unsuccessful. On account of the exceedingly poisonous
and corrosive character of anhydrous hydrofluoric acid, which was
involved in the experiments, more than one chemist lost his life.
If platinum vessels were used, a chocolate-coloured powder, PtF4,
was obtained, and carbon vessels were attacked, with the formation
of a gaseous fluoride, CF4. Attempts to electrolyse hydrofluoric
acid met with no success ; if the aqueous acid was used, only

XXII

THE HALOGENS

417

oxygen and hydrogen were obtained, whilst the anhydrous acid is
a non-conductor of electricity.

It was not until 1886 that Moissan, after a long series of unsuc-
cessful attempts, was able to bring free fluorine to the light of
day ; his triumph was the culmination of a long, dangerous, and
expensive research. Moissan 's success dated from his discovery
that anhydrous hydrofluoric acid, although an insulator, became
an electrolyte when potassium hydrogen fluoride, KHF2, was
dissolved in it. If this solution was electrolysed in a U-tube com-
posed of an alloy of platinum and iridium, with electrodes of the
same material, the whole being strongly cooled to minimise the
corrosive action
of the electrolyte,
then hydrogen was
evolved from the
cathode, and the
long-sought ele-
ment fluorine was
given off from the
anode as a gas.
In 1899 Moissan
found that the ex-
pensive platinum
apparatus could
be replaced by one
of copper, which
apparently be-
comes coated with
a protecting film of
fluoride. The elec-
trodes, however,
must still be of
platinum-iridium.
The apparatus
is shown in Fig.
210. On the left
is the U-tube, of
300 c.c. capacity, containing 60 gm. of acid potassium fluoride
dissolved in 200 c.c. of anhydrous hydrofluoric acid. The electrodes
are insulated by stoppers of fluorspar, covered outside with shellac
(Fig. 211). The tube is immersed in a bath of methyl chloride,
b.-pt. — 23°, which is constantly renewed, and a potential of 50 volts
applied. Hydrogen is evolved from the cathode ; the fluorine
coming from the anode, at the rate of about 5 litres per hour, is
passed through a platinum or copper spiral, cooled in methyl
chloride, and a tube of the same metal packed with fused sodium

E E

FIG. 211.— U-Tube in Moissan's Apparatus.

418 INORGANIC CHEMISTRY CHAP.

fluoride, to remove hydrofluoric acid. By collecting- and measuring
the hydrogen, and absorbing the fluorine in iron wire in a weighed
tube, Moissan found that for every gram of hydrogen evolved from
the cathode the iron increased in weight by 19 grams. The gas
was therefore free fluorine. According to Moissan, the electrolyte
is really potassium fluoride, the anhydrous hydrofluoric acid acting
as an ionising solvent :

Fluorine has also been prepared by the electrolysis of fused
KHF2 or NaHF2, in a copper vessel serving as the cathode ; the
anode was a graphite rod enclosed in a permeable diaphragm.
Brauner (1894) also obtained small quantities of fluorine by heating
potassium fluorplumbate, PbF4,3KF,HF, procured by the action
of hydrofluoric acid on potassium plumbate (p. 926). At 230-250°
this loses hydrofluoric acid ; at higher temperatures free fluorine is
evolved :

PbF4,3KF,HF = HF + PbF4,3KF ;

PbF4,3KF.= PbF2,3KF + F2.

Properties of fluorine. — Fluorine is a pale greenish-yellow gas,
which, when once prepared, has very little action on glass below
100°, and may be kept in glass vessels. It has a powerful odour,
resembling hypochlorous acid, but is not so poisonous as hydro-
fluoric acid vapour. By weighing the gas in a glass flask, Moissan
found the density 19-21 (H = 1), from which the formula F2
follows. The gas was liquefied in 1897 by Moissan and Dewar,
who cooled it in liquid air boiling in a vacuum. It forms a clear
yellow liquid, b.-pt. — 187°, sp. gr. 1-14. By cooling in liquid
hydrogen, Dewaf (1903) obtained solid fluorine, m.-pt. — 233°,
also pale yellow in colour, but becoming colourless at — 252°.

Fluorine fumes in moist air, forming hydrofluoric acid and con-
siderable amounts of ozone : F2 -f H2O = 2HF + O. Fluorine
is the most active element known ; it forms no compounds with
oxygen and chlorine, but combines with nearly every other element.
It combines with bromine and iodine, forming BrF3 and IF5, both
colourless liquids.

Fluorine unites with hydrogen explosively even at — 252° ;
sulphur, selenium, tellurium, phosphorus, iodine, bromine, arsenic,
antimony, silicon, boron, carbon, and potassium all ignite spon-
taneously in the gas, and burn with the formation of fluorides. A
jet of fluorine ignites at once in a jar of hydrogen, burning with^a
red-bordered flame, and producing HF, which attacks the glass
jar. Lead and iron are rapidly attacked ; magnesium, manganese,
nickel, aluminium, and silver take fire when gently warmed. Gold
and platinum are not attacked at the ordinary temperature, but

xxn THE HALOGENS 419

are corroded and form fluorides on heating. Alcohol, ether, and
turpentine take fire spontaneously in the gas. Potassium chloride
is decomposed with evolution of chlorine :

2KC1 + F2 = 2KF + C12.

By the action of fluorine on aqueous alkalies, hydrogen peroxide is
formed. Fluorine can replace oxygen in many acids without pro-
ducing much change in chemical properties, e.g., it forms fluoriodates,
MF2IO2,IF3(OH)3, and replaces oxygen in niobates and tantalates.
The two elements are both strongly electronegative, and for this
reason do not form compounds with each other. Since a mixture
of fluorine and oxygen explodes when subjected to the silent dis-
charge, it has been assumed that the element can combine with
ozone.

Hydrofluoric acid, HF. — Hydrogen and fluorine combine with
explosion under all conditions, forming hydrogen fluoride, or hydro-
fluoric acid, HF. This is more conveniently obtained by the action
of acids on fluorides, or by
heating acid potassium fluoride :
KHF2 = KF + HF.

If powdered fluorspar is dis-
tilled with concentrated sulphuric
acid in a lead retort, connected
with a lead receiver containing
water (Fig. 212), the vapour of
hydrofluoric acid dissolves in the

latter to form a Colourless, FlG- 212.— Lead Retort for preparation of
, P . i. . i T . . Hydrofluoric Acid.

strongly-turning liquid, which is

a solution of hydrofluoric acid : CaF2 -f H2SO4 = CaS04 -f 2HF.
This attacks glass strongly, and is kept in lead or gutta-percha
bottles. It is used for etching or engraving on glass. The latter
consists of silicates of the metals of the alkalies and alkaline earths ;
the hydrofluoric acid removes the silica in the form of silicon
fluoride : Si02 -f 4HF = SiF4 + 2H20. Etchings made with the
liquid acid are clear ; those made with the gas, or a mixture of
aqueous acid and ammonium fluoride, are opaque.

EXPT. 172. — A watch-glass is covered with beeswax by melting
the latter on it and draining off the superfluous liquid. When the wax
has hardened, a device is scratched through with a needle, and the
glass placed over a lead dish containing a mixture of powdered fluorspar
and concentrated sulphuric acid. The parts of the glass exposed will
be found to be etched if the wax is removed after a few minutes by
warming the glass.

The commercial acid contains about 40 per cent, of HF ; its
sp. gr. is 1 -130. It is used for glass etching, for removing silica from

E E 2

420

INORGANIC CHEMISTRY

canes, and as an antiseptic. The so-called " wild yeasts," which
produce fusel oil in fermentation, are killed by small quantities of
fluorides, whilst normal yeast-cells may be accustomed to the
latter. Lactic and butyric fermentations are also inhibited.
Sodium fluoride may be used for this purpose. Zinc and sodium
fluorides are used in preserving wood.

If aqueous hydrofluoric acid is neutralised with caustic potash,
and the liquid evaporated in a platinum dish, cubic crystals of
potassium fluoride, KF, are obtained. If, to the neutralised liquid,
a further equal volume of hydrofluoric acid is added and the liquid
evaporated in a platinum dish, crystals of potassium hydrogen
fluoride, KHF2, or KF, HF, called Fremy's salt, are obtained. This
may be dried by heating, and is relatively stable. If it is heated

in a platinum
retort, con-
nected with a
platinum con-
denser cooled
by a f reez
ing mixture,
anhydrous hy-
drofluoric acid
distils over
(F i g. 213).
The acid
fluoride on
heating forms
the normal

salt : KHF2 = KF + HF. The anhydrous acid, first prepared
in this way by Fremy in 1856, may also be obtained by heating
lead fluoride in hydrogen : PbF2 -f H2 = Pb -f 2HF.

Traces of moisture may be removed from hydrofluoric acid by
electrolysis with platinum electrodes, when, as long as water is present,
ozonised oxygen is evolved. When all the water is removed, the acid
becomes non-conducting.

Anhydrous hydrofluoric acid is a colourless, strongly -fuming
liquid, sp. gr. 0-988, boiling at 194° ; it should therefore be kept
in a freezing mixture. It does not solidify until cooled to — 102° ;
the transparent, colourless solid melts at — 92-3°. When quite
free from water it does not attack glass or metals at the ordinary
temperature, except potassium, which explodes in contact with the
acid. In the presence of traces of water, the acid attacks glass
violently, and dissolves most metals with evolution of hydrogen :
Fe + 2HF = FeF2 -f H2. The noble metals are not attacked,
but gutta-percha (which resists the aqueous acid) and most organic

FIG. 213. — Platinum Retort and Condenser for preparing
Anhydrous Hydrofluoric Acid.

xxn THE HALOGENS 421

materials are rapidly corroded. The acid and its vapour are
dangerous corrosive poisons.

Aqueous hydrofluoric acid forms an acid of maximum boiling
point, 120°, containing 36 per cent, of HF.

The composition of hydrofluoric acid was determined by Gore
(1869), who heated silver fluoride at 100° in hydrogen in a platinum
vessel, and obtained twice the volume of hydrofluoric acid gas.
The formula at 100° is therefore HF : 2AgF + H2 = 2HF + 2Ag.
Mallet (1881), by weighing the vapour at 30-5° in a glass flask
coated inside with paraffin wax, obtained the density 19-66, corre-
sponding with the formula H2F2. Thorpe and Hambly (1889)
showed, by determining the vapour densities in a platinum flask
at various temperatures and pressures, that the gas is associated,
the density varying considerably with the temperature and pressure.
At 88° and 741 mm. the molecular weight corresponds with HF ;
at lower temperatures it approximated to H3F3. No indication
was found of the separate existence of H2F2, the density falling off
continuously with rise of temperature, or diminution of pressure,
to the limiting value corresponding with HF.

In concentrated solutions the acid appears, from freezing-point
measurements, to be H2F2 ; in dilute solutions it has the formula
HF.

The fluorides differ in many respects from the other halogen com-
pounds. Silver fluoride is very soluble in water ; calcium fluoride
is nearly insoluble. The iron compound corresponding with cryolite,
viz., FeF3, 3NaF, is insoluble. If a standard solution of a ferric salt
is added to a solution of sodium fluoride, this compound is precipi-
tated, and if a little ammonium thiocyanate is added, the excess of
ferric salt gives a red colour. Fluorides may be titrated in this
way.

The fluorides also readily form complex and double salts with
hydrofluoric acid : e.g., HBF4, H2SiF6, H2NbOF5, etc. The acid
fluorides, such as KHF2, may be regarded as derived from H2F2,
which behaves as a dibasic acid.

The strength of hydrofluoric acid. — The heat of neutralisation of
a strong acid by a strong base is always approximately the same,
and equal to 13-7 kg. cal. (p. 295), this being the heat evolved in
the reaction : H'Aq. -f- OH'Aq. = H20. Hydrofluoric acid, how-
ever, on neutralisation evolves 16-3 kg. cal., whilst if excess of the
acid is added to the neutral salt, 0 -3 kg. cal. is absorbed. Measure-
ments of the conductivities of solutions of the acid show that it is
much less ionised than the other halogen hydracids ; in decinormal
solutions the percentage ionisations of hydrofluoric and hydro-
chloric acids are 15 and 92, respectively. On neutralisation,
the un -ionised molecules break up into ions as the reaction
H' -f OH' = H2O proceeds, and the abnormally large heat of

422

INORGANIC CHEMISTRY

CHAP.

neutralisation shows that heat is evolved in the reaction : HF *-.
H* -f F'. The absorption of heat on adding excess of acid is no
doubt due to the formation of acid salts, e.g., KHF2.

The weak acetic acid has a nearly normal heat of neutralisation,
13-3 kg. cal. ; hypochlorous acid has a very small heat of neutralisa-
tion, 9-8 kg. cal., since it is unable to neutralise an alkali in solution,
on account of hydrolysis : NaOCl ^ NaOH + HOC1. Hydro-
chloric, hydrobromic, and hydriodic acids are about 92 per cent,
ionised in decinormal solution ; chloric and perchloric acids are
almost as strong, whilst hypochlorous acid is only 0-02 per cent,
ionised. Carbonic acid is 0-17 per cent, ionised, so that it is able to
displace hypochlorous acid from its salts when the latter are exposed
to air.

Atomic weight of fluorine. — Older determinations of the atomic
weight of fluorine were based on the reaction CaF2 -f- H2SO4 =
CaS04 + 2HF. Since it is difficult to carry this to completion,
somewhat varying results were found, the mean of good deter-
minations being 18-85 (H = 1). By converting pure lime into
calcium fluoride by treatment with hydrofluoric acid, the ratio
CaO : CaF : : 1 : 1-3932 was found. Thus :

'• F =19-01 <0 = 16), or 18-9 (H=l).

The halogens. — The elements fluorine, chlorine, bromine, and
iodine are so obviously related in their chemical properties as to
lead to their separation from the remaining elements to form a
group, or family, which is called the halogen group (Greek hols =
sea-salt). If we consider the properties of the free elements of the
halogen group, and of their compounds, a marked gradation in the
order given above is apparent. This is seen, in the first place,
in the physical properties of the elements: —

Atomic
Element, weight.

Fluorine 18-9

Chlorine 35-18

Bromine 79-29 liquid

Iodine 125-91

Physical
state. Colour.

pale

greenish-
yellow
greenish -
yellow
(liquid
yellow)
dark red
(vapour
red)
black
(vapour
violet)

Melting Boiling Sp. gr. of Solubility
point. point. liquid, in water

atO°
decom-

-233° -187°

1-14

gas

-102° -33-6° 1-55

-7-3°

63°

3-19

solid

113° 184-4° 5 (solid)

poses

9-92 gm.
per litre
at 10°

41-5 gm.
per litre

0-3 gm.
per litre.

xxn THE HALOGENS 423

In a similar way, we may compare the physical properties of the
hydrogen compounds, all of which are acids :

Compound Melting Boiling Density of Heat of formation

point. point. liquid. in kg. cal.

HF (polymerised)- 92-3° 19-4° 0-988/15°

HC1 -112-5° -83° 0-929/0° 22

HBr - 86° -68-7° 1-78 12-1

HI - 51-3° -36-7° - 6-1 (from

solid iodine)

The physical properties of hydrofluoric acid are seen to be abnormal ;
this results from the circumstance that it is associated even in
the gaseous state below 80°, forming HnFn, whereas the other
substances are normal. Polymerisation invariably leads to an increase
of melting and boiling points. The abnormal ionisation of hydrofluoric
acid has already been described.

The stability of the hydrogen compounds, as measured by their
dissociation on heating, is in the order HF > HC1 > HBr > HI,
i.e., in the or der of the heats of formation. Thus, hydrogen iodide is
appreciably dissociated at 360°, but hydrogen chloride only slightly
at a white heat. The halogens also displace one another from their
binary salts in the order of the heats of formation, viz. : F -» Cl
-> Br -> I. In the oxygen compounds, however, iodine displaces
chlorine (p. 413) : 2KC1O3 + Ia = 2KIO3 + C12. The relation of
the stabilities of the oxygen compounds, from fluorine (no oxide) to
iodine (stable I2O5), to the heats of formation has already been
considered (p. 391).

EXERCISES ON CHAPTER XXII

1. Describe the methods used for the preparation of bromine in
the laboratory and in industry. What is the action of the element
on (a) cold dilute caustic potash, (6) hot concentrated caustic potash,
(c) mercuric oxide, (d) silver nitrate ?

2. How is bromic acid prepared ? What are its properties, and those
of its salts ? How may a bromate be distinguished from a chlorate
and an iodate ?

3. Describe the methods used in the manufacture of iodine. What
impurities does the commercial substance contain, and how may they
be separated ?

4. What is the action of heat on (a) bromine, (b) iodine, (c) hydriodic
acid, (d) barium bromate, (e) barium iodate ?

5. How are the oxides of iodine prepared ? What formulae are usually
given to the oxy-compounds of iodine ?

6. Starting from iodine, how would you prepare (a) hydriodic acid,
(6) iodic acid, (c) periodic acid ? Describe briefly the properties of

424 INORGANIC CHEMISTRY CH. xxn

these substances, and compare them with the corresponding compounds
of chlorine.

7. In what forms do bromine, iodine, and fluorine occur in Nature ?
How is hydrofluoric acid prepared from fluorspar ?

8. How is anhydrous hydrofluoric acid prepared ? By what means
did Moissan prepare fluorine from this substance ?

9. Describe the properties of fluorine, and explain why this element
is included in the series of halogens.

CHAPTER XXIII

ATOMIC HEATS AND ISOMORPHISM

The determination of atomic weights. — The methods used in
deciding which multiple of the equivalent of an element is the atomic
weight have already been referred to briefly (p. 146). They
include :

1. Determination of the least weight of the element found in the
molecular weights of volatile compounds, the molecular weights being
found from the vapour densities by Avogadro's law (p. 143) ; this
requires that the element shall form a number of volatile compounds,
which is not always the case.

2. The molecular weights of compounds may be determined in
solution by the freezing-point, boiling-point, or vapour -pressure methods
(Chapter XXI).

3. The Atomic Heat method, applicable to solid elements, especially
metals.

4. Isomorphism.

5. Chemical analogies with compounds of elements of known atomic
weight.

6. The ratio of the specific heats of gases (pp. 146, 598).

7. The Periodic Law (Chapter XXIV).

The application of as many of these methods as possible gives
a valuable check on the atomic weight. Thus, if the atomic weight
has been fixed approximately from the specific heat, the vapour
density of one volatile compound may be most valuable in con-
firmation, although alone it could not have given certain results,
since then it could not be assumed that the compound contained

tDnly one atom of the element.
Atomic heats, and isomorphism, will be considered in the present
chapter ; the Periodic Law is discussed in the following chapter.
[

ATOMIC HEATS.

Dulong and Petit's Law.— P. L. Dulong and A. T. Petit, in 1819,
discovered a very simple relation between the atomic weights and
specific heats of solid elements, viz., that the product of the two, which

425

426

INORGANIC CHEMISTRY

CHAP.

they called the atomic heat, is constant. Dulong and Petit's law asserts
that the atomic heats of solid elements are constant, and approximately equal
to 6'3.

Quantities of solid elements in the proportion of their atomic
weights are therefore raised through 1° in temperature by identical
quantities of heat. The heat capacity of a solid element is a pro-
perty of its atoms : Dulong and Petit expressed their result in the
statement : the atoms of all solid elements have the same capacity for
heat. By assuming that half the energy of a monatomic solid,
due to atomic vibration, is kinetic, and half potential (as in the
vibrations of a pendulum), and that the kinetic energies of the atom
of the solid and that of a monatomic gas are equal at the same tem-
perature, Boltzmann (1871) arrived at the result that the atomic
heat of the solid is double that of the monatomic gas, viz., 2x3 = 6
cal. (p. 598). The following table gives the results determined near
atmospheric temperature.

TABLE OF ATOMIC HEATS.

Atomic heat —

Atomic weight Specific heat

Atomic weight

Element.

(H = 1)

(20° to 100°)

x Specific heat.

Arsenic

74-5

0-0827

6-16

Bismuth

206-5

0-0303

6-27

Bromine (solid) ...

79-3

0-084

6-65

Calcium

0-17

6-80

Cobalt

58-5

0-1030

6-05

Copper

0-0936

5-90

Gold

196

0-0316

6-20

Iodine

126

0-054

6-80

Iron

55-5

0-1146

6-35

Lead

206

0-0314

6-52

Lithium

6-9

0-94

6-48

Magnesium

0-2492

5-95

Mercury (solid) ...

199

0-0335

6-66

Nickel

0-1092

6-32

Phosphorus (yellow)

0-1981

6-15

Platinum

194

0-0320

6-20

Silver

107

0-0566

6-06

Sulphur

0-1780

5-70

Tin

• 118

0-0556

6-55

Uranium

236

0-0276

6-50

Zinc

0-0931

6-05

Mean atomic heat = 6-30

In order to obtain agreement with the law, Dulong and Petit
found it necessary to alter some of the atomic weights current at the

xxm ATOMIC HEATS AND ISOMORPHISM "427

time : with one or two exceptions these modifications have been
confirmed.

The exceptions to Dulong and Petit 's law, which all give atomic
heats lower than 6-3, occur among elements of low atomic weight
and high melting point. Thus, although lithium (at. wt. 7 ; m.-pt.
180°) and sodium (at. wt. 23 ; m.-pt. 97-6°) conform to the law, the
following elements, with atomic weights lower than 30, all have
atomic heats considerably below 6-3 :

Melting Atomic Specific Atomic

Element. point. weight. heat at 15° heat at 15°

Beryllium ... 1300° 9 0-3756 3-4

Boron (cryst.) above 2000° 11 0-239 2-64

Carbon (diamond) do. 12 0-113 1-35

„ (graphite) do. 12 0-160 1-92

Silicon (cryst.) c. 1200° 28 0-170 4-75

Weber (1875), however, found that the specific heats of boron,
carbon, and silicon increase fairly rapidly with the temperature at
which the determination is carried out, and the same result was
found for beryllium by Humpidge (1885).

Diamond. Graphite. Boron. Silicon. Beryllium.

°C. At ht. °C. At ht. °C. At. ht. °C. At. ht. °C. At. ht.
- 50 0-76 - 50 1-37 - 40 2-11 - 40 3-81 0 3'42

10-7

1-35

10-

1-92

26-6

2-62

•6 4

•75

100

4-28

58-3

1-84

61-

2-39

76-7

3-01

•32

200

4-93

140

2-66

201-

3-56

177-2

3-63

184

•3 5

•63

300

5-38

247

3-63

249-

3-90

233-2

4-33

232

•4 5

•68

400

5-61

615

5-33

640

5-40

iM.

500

5-65

808

5-44

832

5-42

980 5-47 980 5-63

At high temperatures, the atomic heats of these elements approach
the normal value, 6-3. The variation with temperature is shown
by the curves of Fig. 214. The atomic heats of other elements,
which have the normal value 6-3 at the ordinary temperature, also
increase somewhat with the temperature, but not to the same
extent as those with abnormal atomic heats.

Thus, the atomic heats of platinum at 18-100°, and 1230°, are
6-2 and 8 '84, respectively.

Atomic heats at low temperatures.— The fact that the atomic
heats of boron, carbon., silicon, and beryllium become larger,
and approach the normal values, as the temperature rises, sug-
gests that the elements with normal atomic heats may be, at
the ordinary temperature, in a region which is only attained

428-

INORGANIC CHEMISTRY

CHAP.

at higher temperatures by the former elements. In this case
the atomic heats of the elements which behave normally should
become abnormally small at low temperatures. This has been
found by experiment to be the case. The atomic heats of all solid

6-3 ~ ~

6 --

:::::::::::::::::::;§;--:::::

- ^-i Jeryll iumL- --Grap litie

::::::::::::^!:::::|^!::::::

i;ji:si!!= = = = H!?^j-S

.2?

^, "

"IrnnA ^ ^- " =C -

S- -+

Jeryllium-9- """ZII*~i-I"

•£

___J l\*-t-~?Z

:± — HQ/T^-^--

S A n'

Z T:

7" ~Z

,- . . A " 3!

/ -

rpo .,

amond * '

±_: ::::

-200°

0° 200° 4OO°

600° 800° 1000"

Temp. °C.
FIG. 214. — Atomic Heat Curves.

elements fall to small values at low temperatures, some more rapidly
than others, and it is probable that at the absolute zero, — 273°,
the atomic heats are all zero. In the case of the diamond, the atomic
heat is actually zero at temperatures below — 230°.

Element.

Carbon

Aluminium

Silicon

Iron

Copper

Zinc

Silver ...

Lead

Atomic heat

+20° to 100°

2-4

5-9

5-2

6-4

6-2

6-1
,. • 6-1

6-4

Atomic heat

- 188° to + 20(
1-15
4-73
3-34
4-80
4-88
5-53
5-51
6-21

Atomic heat
-253° to -195°
0-03
M2

0-77
0-98
1-22
2-52
2-62
4-96

The values for the diamond at low temperatures are :
Temperature °C.... 896 85 - 41 - 64 - 181 - 231

Atomic heat

5-45 2-12 0-86 0-66 0-03

- 243
0-00 0-00

The dependence of atomic heat on temperature is shown for a few
elements in the curves of Fig. 215, from the experiments of Nernst.

xxin ATOMIC HEATS AND ISOMORPHISM 429

The following results were obtained by Kamerlingh Onnes and
Keesom (1915), at the temperatures of liquid hydrogen :

Lead. Copper.

Temp. abs. Atomic heat. Temp. abs. Atomic heat.

14-19° 1-56 15-24° 0-05

22-31° 2-98 21-505° 0-14

46-25° 5-04

The quantum theory. — The rapid diminution of the specific heats of
solids at low temperatures, and the convergence to zero in the neighbour-
hood of the absolute zero, is in agreement with the theory of energy
quanta, due to Planck (1906). According to this theory, the atoms of a
solid do not take up heat energy continuously, but in finite amounts,
called quanta, which may be considered as atoms of energy. The value

T= 50 100 150 200 250 300 350 400
Fia. 215. — Atomic Heats at Low Temperatures.

of the quantum, e, varies from element to element, and is equal
to hv9 where h is a universal constant, equal to 6-55 X 10~2r,
known as Planck's constant, and v is the atomic frequency, characteristic
of each element. In the case of sodium, for instance hv = € =
(6-55 X 10~27) x (5-01 x 1014) = 3-28 X 10 ~12 ergs, which is about one-
sixtieth the kinetic energy of a hydrogen molecule at 0°. This value of
v is the frequency of the light emitted by incandescent sodium vapour.
The " deviations " from Dulong and Pe tit's law at low temperatures
are explained by the theory of energy quanta ; the former law is a
limiting case of a more general law deducible from the new theory.
According to this, the atomic heat of a monatomic solid element is given
by the expression, due to Einstein (1907) :

x2 ex

Atomic heat = 3R -r— -, v2 >

(e — l)

rhere x = ftv/T, the constant /3 being equal to Planck's constant

430 INORGANIC CHEMISTRY CHAP.

h divided by the gas constant, R, in absolute units (p. 149) and multi-
plied by Avogadro's constant, N (p. 268) :

AN _(6-55 X IP'2?) X (6 X 1023)

P = ^r; „ » . •.— ?-/vr ~ 4-8 X 10 ".

Planck's theory leads to the assumption of the atomic structure of
radiation : this is also made up of quanta, hv, where v is the frequency.
The " critical energy " of a molecule (p. 354) also appears to be of the
form hv, where v is the frequency of some type of radiation absorbed
by the substance.

According to Debije, the atomic heats at very low temperatures
should be proportional to the cube of the absolute temperature :
At. ht. = kT\ This has been confirmed by Nernst and others.

Atomic weights from specific heats. — Dulong and Petit 's law
obviously gives an approximate value of the atomic weight of a
solid element if the specific heat is known :

Atomic weight = 6-3 -r Specific heat.

Thus, an analysis of a volatile chloride of uranium shov/s that it
has the following percentage composition :

Uranium 62-66
Chlorine 37-34

100-00

The equivalent of uranium, or the weight combining with 35-2

35-2
parts of chlorine, is 62-66 X 37734 = 59*1. The vapour density

of the chloride was found by Zimmermann to be 193 (H = 1), hence
the approximate molecular weight is 191 X 2 = 382. This will

382
contain 37-34 X JQQ = 142-5 parts, which is nearly equal to

4 X 35-2 = 140-8 parts, or four atoms, of chlorine. The formula
of the chloride is therefore Ua;Cl4, where x = 1, 2, 3, 4 . . . etc.

The weight of uranium in a molecular weight of the chloride is,
approximately, 382 — 142-5 = 239-5. But this is very nearly
equal to 4 X 59-1 = 236-4, i.e., four times the accurately determined
equivalent. Thus, U^ = 236-4. It has still to be decided whether
this is the atomic weight of uranium, or a multiple of it. Thus, the
following formulae of the chloride are possible :

At. wt. of

Formula. Uranium.

UC14 236-4

U2C14 118-2

U3C14 78-8

U4C14 59-1

xxm ATOMIC HEATS AND ISOMORPHISM 431

To decide which of these is the correct formula, an approximate
value of the atomic weight of uranium must be found. The specific
heat of solid metallic uranium is 0-0276; hence, by Dulong and
Petit's law, the atomic weight is approximately 6-3/0-0276 = 228.
This shows that the exact value is 2364, and hence the formula of
the chloride is UC14.

It must be carefully noticed that the value of the atomic weight
deduced from Dulong and Petit's law is approximate only, and is used
to decide on a particular multiple of the exact equivalent. The mole-
cular weight of the compound is also found from the exact chemical
analysis.

The more exact expressions for the atomic heat derived from the
quantum theory lead to more accurate values of the atomic weights,
but not so accurate as those found by exact chemical analysis.

Molecular heat of a compound. — An extension of Dulong and
Petit's law to solid compounds was made by F. Neumann in 1831.
He found that the specific heats of solid substances of similar composition
are inversely proportional to their molecular weights. Thus :

Molecular Specific Molecular

Substance- weight. heat. heat.

Calcium carbonate, CaCO3 ... 100 0-2044 20-44

Magnesium carbonate, MgCO3 84 0-2270 19-1

Ferrous carbonate, FeCO3 ... 116 0-1819 21-1

Zinc carbonate, ZnCO3 ... 125 0-1712 21-4

Barium carbonate, BaCO3 ... 196 0-108 21-1

Lead carbonate, PbCO3 ... 266 0-081 21-6

The molecular heat of a solid compound is the product of its specific
heat and its molecular weight. Neumann's law shows that the
molecular heats of similar compounds are alike. The molecular
heats of the carbonates of the alkaline-earth metals, etc., of
the general formula RC03, are 20 (approximately) ; the
sulphates, RS04, of the same metals have a molecular heat of
about 25.

The relation between Neumann's law and that of Dulong and
Petit was pointed out by Joule hi 1844. Joule's law (often called
Woestyn's law) states that the molecular heat of a solid compound is the
sum of the atomic heats of its constituents.

This was confirmed by the experiments of Kopp (1865). It
indicates that the atomic heat of an element is unchanged by com-
bination, or the molecular heat of a solid compound is additively
composed of the atomic heats of its elements. The heat content

432 INORGANIC CHEMISTRY CHAP.

of any solid, therefore, seems to reside in its atoms. With gases,
the case is quite different (p. 598), since the kinetic energy of the
molecule, is predominant.

An example of Joule's law is the calculation of the molecular heat of
silver iodide :

Atomic heat of silver =107x0-057 6-10

Atomic heat of iodine = 126 X 0-054 = 6-80

Sum of atomic heats =6-1+6-8 =12-9

Molecular heat of silver iodide* = molecular

weight X specific heat = 233 X 0-061 = 14-2
The molecular heats of lead bromide and iodide may also be calculated
from the sums of the atomic heat of lead and twice the atomic heat of the
halogen :

PbBr2 = 6-48 + 2 X 6-65 = 19-78
Pb I2 = 6-48 + 2 X 6-80 = 20-08
(Pb 4- 2 Br) x sp. ht. of lead bromide = ('206 + 2 x 79-3) x 0-0533

= 19-4
(Pb -f 21) x sp. ht. of lead iodide = (206 + 2 x 126) x 0-0427 =

19-6
The agreement, as will be seen, is only approximate.

It is possible to calculate by means of Joule's law the atomic
heats of elements in the solid state in cases where these cannot be
directly determined. Thus, the atomic heat of solid chlorine may be
calculated as follows :

Specific heat of silver chloride = 0-091 .'. molecular heat of AgCl
= 0-091 X (107 + 35-2) = 12-96. This, however, is the sum of the
atomic heats of silver and of solid chlorine ; hence :
atomic heat of solid chlorine = molecular heat of silver chloride -
atomic heat of silver = 12-96 — 6-10 = 6-86.

From the molecular heats of their compounds, Kopp deduced the
following atomic heats :

Boron 2-7 Phosphorus 5-4

Carbon 1-8 Sulphur 54

Silicon 4

These agree quite well with the values determined directly at
0°-100°, although they are all abnormal. The abnormal atomio
heats are therefore preserved in combination. The calculated
values for solid oxygen, nitrogen, and fluorine are also abnormal, as
would be expected since these elements are non-metals of low atomic
weights.

XXITI

ATOMIC HEATS AND ISOMORPHISM

433

11-9

X 15-88)
calcium and carbon =
oxygen = £(20-4- 8-0)

Molecular heat of calcium carbonate = (39'8
X 0-206 = 20-4. Sum of atomic heats of
6-8 + 1-8 = 8-6 .'. atomic heat of solid
= 3-9.

Molecular heat of ice = 18 X 0*92 = 16-5; atomic heat of solid
oxygen = 3-9 .*. atomic heat of solid hydrogen — ^ (16-5 — 3-9)
= 6-3, which is the normal value.

The extension of the quantum theory to compounds is still incom-
plete. Nernst suggests that the energy of the molecules themselves
can be calculated from Debije's T3 formula, whilst that of the atoms in
them follows Einstein's law.

CRYSTALLOGRAPHY

Crystals. — A distinction has already been drawn between crystal-
line and amorphous substances. The most obvious difference
between the two is that of external form : whereas amorphous solids
are found in irregularly-shaped pieces, crystals usually have definite
shapes, by which they are recognised. Another difference is in the
fracture : crystals break into more or less similarly-shaped pieces,
with plane faces meeting in sharp edges, whilst amorphous solids,
such as glass or pitch, break into very irregular
pieces, showing curved faces, with concentric
rings, such as are seen inside an oyster-shell.
These two kinds of fracture are known as crystalline
fracture and conchoidal fracture, respectively.

A crystalline substance may, however, be
recognised even if in powder, and with no
apparent external form. With the exception of
crystals of the regular system (see below), all
fragments of crystals act upon polarised light, and
if the powder is examined under a microscope
so arranged that the light passes through a pair of
crossed Nicol prisms, and is therefore totally
extinguished, it is found that light passes through
the crystal grains, which are seen beautifully
coloured on a dark ground. Again, if a crystal
of gypsum is touched with a red-hot needle
on one of its faces, a white patch of anhy-
drous calcium sulphate develops (Fig. 216) :
CaS04,2H20 = CaSO4 + 2H2O. This patch is not circular, but
elliptical, showing that the heat is conducted through the crystal
more readily in one direction than in the perpendicular direction.

We thus are able to recognise some definite internal arrangement,
or internal structure, in the crystal, and the outer form is determined
by this structure. Even if the outer form is destroyed by breaking,

F F

ElG. 216. — Crystal
of Gypsum show-
ing Plane of Sym-
metry.

434

INORGANIC CHEMISTRY

CHAP.

or grinding, the internal structure corresponding with it remains, and
may be recognised. If, however, the above experiments are tried
with a piece of glass, which is an amorphous solid, it is found that
no light passes under crossed Nicols, and if the glass is coated with
paraffin wax, the latter is melted in a circular patch when a hot
needle is pressed upon the solid. The results are the same even if the
glass has been cut into any external form like that of a crystal :
the resemblance to a crystal is spurious, and the glass remains
all the time an amorphous body. The internal structure is more
important than the external form.

The definite internal structure of crystals is almost certainly due
to some definite or ordered arrangement of the atoms or molecules in
the crystal : this arrangement can be detected by the reflection of
X-rays from the crystal faces (p. 1018). The molecular structure is
found to be symmetrical, i.e., a definite pattern will be repeated over
and over again in definite directions in space, in the same way, for

FIG. 217.— Cube.

FIG. 218.— Octahedron.

FIG. 219.— Combination of
Cube and Octahedron.

instance, as the pattern of a wall-paper. To the internal symmetry
of the arrangement of the atoms or molecules there also corresponds
an external symmetry of the crystal form.

Symmetry of crystals. — The symmetry of a crystal form is deter-
mined by regularities in the positions of the similar faces, edges, etc.,
of the crystal. A crystal having all its faces alike is termed a simple
form. Thus, both the cube in Fig. 217 and the octahedron iri Fig. 218
are simple forms, because all the faces of the first are identical
squares, and all those of the second are identical equilateral triangles.
A crystal having sets of faces corresponding with two or more simple
forms is called a combination form. Thus, the crystal of galena
(PbS) shown in Fig. 219 is a combination of the cube and the
octahedron, since it contains sets of faces derived from each.

The regularities in the positions of faces, edges, etc., i.e., the
symmetry of the crystal, are defined in terms of planes of symmetry,
axes of symmetry, and a centre of symmetry. A plane of symmetry
divides a crystal into two similar and similarly-placed halves, each

ATOMIC HEATS AND ISOMORPHISM

435

FIG. 220. — Axis of
Symmetry of Cube.

being the mirror-image of the other. Thus, a crystal of gypsum
is divided by the plane shown in Fig. 216 into two similar and simi-
larly-placed halves ; this is the only plane of
symmetry possessed by the gypsum crystal.
An axis of symmetry is denned as an axis such
that, if the crystal is rotated around it, the
crystal occupies the same position in space more
than once in a complete turn. Thus, the axis
shown in Fig. 220 is an axis of fourfold
symmetry, since the cube takes up the same
position in space four times on rotation
through 360° about this axis. Axes of two-,
three-, four-, and six-fold symmetry occur, when
the crystal comes to occupy the same posi-
tion in space 2, 3, 4, or 6 times in a complete
revolution, i.e., on rotation through 180°, 120°,
90°, or 60°.

A crystal has a centre of symmetry when like
faces are arranged in pairs in corresponding
positions on opposite sides of a central point.

An examination of a cube shows that it possesses 9 planes of
symmetry (Fig. 221) ; it has 13 axes of symmetry (3 of fourfold, 4
of threefold, and 6 of twofold symmetry), and a centre of symmetry.
It is therefore said to possess 23 elements of symmetry, which is the
highest number possible in a crystal. Some crystals have no plane
of symmetry, others have no axes of symmetry, others have no
centre of symmetry, and a few have no element of symmetry at all.
The crystallographic symmetry depends on the internal molecular
structure, and need not correspond with the geometrical symmetry
except in the perfect crystal, since the crystal may have certain

faces developed to a
greater extent than
others. The angles
between the faces, how-
ever, are the same
both in the ideal
crystal and in the
actual, distorted, crys-
tals, and these angles
are the important
measurements in de-
termining the crystal
form. Thus, the angles
between the faces of the perfect and distorted octahedra in Fig . 222
are identical.

Crystallographic systems. — It has been shown mathematically

F F 2

FIG. 221.— Planes of Sym-
metry of Cube.

FIG. 222.— Ideal and Dis-
torted Octahedra, show-
ing Constancy of Angles
between the Faces.

436

INORGANIC CHEMISTRY

CHAP.

that thirty-two types of symmetry, differing in the degree
and nature of the elements of symmetry, are possible among

FIG. 223. — Triakisoctahedron
(Three -faced Octahedron).

PIG. 225. — Hexakisoctahedron
FIG. 224.— Icositetrahedron. (Six-faced Octahedron).

crystals, so that the latter may be classified into symmetry groups.
Of these, eleven only include practically all the common
substances.

The usual method of classification, however, is into what are
known as crystal systems. These are related to the crystallographic
axes. The position of any crystal face is denned by the intercepts
made on three axes intersecting in a point inside the crystal. If a
suitable number of axes of symmetry exists, three of them may be
chosen as crystallographic axes, but the latter need not be the axes
of symmetry.

The following types of crystallographic axes occur :
1. Three equal axes at right angles : this corresponds with the
cubic, or regular, system.

Fig. 218 shows the regular octahedron, which is the typical pyramid

form of the regu-
lar system. Fig.
217 is the cube,
which is the
typical prism form
of the system.
The other simple
forms (cf. above)
of the regular
system are the
triakisoctahedron
(Fig. 223), the
icositetrahedron

(Fig. 224), the hexakisoctahedron (Fig. 225), the rhombdodecahedron
(Fig. 226), and the tetrads-hexahedron (Fig. 227). Combinations of
these forms also occur.

FIG. 226. — Rhombdodecahedron.

FIG. 227. — Tetrakis-hexahedron
(Four-faced Cube).

ATOMIC HEATS AND ISOMORPHISM

437

XXIII

Examples of substances occurring in the regular system are :
octahedron : alums, magnetite (Fe3O4), cuprite (Cu2O) ;
cube : fluorspar, common salt, sylvenite (KC1) ;
rhombdodecahedron : garnet ;
tetrakis -hexahedron, etc., iron pyrites (FeS2).

(2) Two equal axes meeting at right angles, and a third, longer or
shorter, axis meeting these at right angles. This constitutes the

FIG. 228.— Tetragonal Bipyramid
First Order.

FIG. 229. — Tetragonal Bi-
pyramid : Second Order.

tetragonal system. Typical pyramid and prism forms are shown in
Figs. 228 — 231. There are two orders of pyramid and prism forms,
according as the horizontal axes terminate at the angles (Figs. 228
and 230), or the middle point (Figs. 229, 231), of the faces.

Zircon (Zr2SiO4), f

potassium dihy-
drogen phosphate
(KH2PO4), and tin-
stone (SnO2) show
prism and pyramid
forms ; potassium
ferrocyanide gives
chiefly pyramid
forms.

If the length of
the vertical axis
(AB) in this (and
other) systems be
denoted by c, and the length of the horizontal axes by a and 6,
with appropriate signs, as shown in Fig. 232, the cubic system may
be denoted by (a a a), and the tetragonal system by (a a c).

FIG. 230. — Tetragonal Prism :
First Order.

^^"

— i —

FIG. 231. — Tetragonal
»rism : Second Order.

438

INORGANIC CHEMISTRY

(3) In the hexagonal system there are four axes, three equal and
intersecting in the same plane at angles of 60°, and a fourth axis,

greater or less than these,
at right angles (a a a c).
Here again there are two
types of pyramid and
prism forms, according as
the lateral axes meet
angles or the mid-points
of faces.

Typical pyramid forms

FIG, 232.— Crystallographic Axes.

FIG. 233.— Hexagonal Bipyramid
First Order.

are shown in Figs. 233 and 234 ; prism forms are shown in
Figs. 235 and • 236. Examples of this form are witherite (BaC03),
beryl, and apatite.

(4) In the rhombic system there are three unequal axes all at right

FIG. 234.— Hexagonal Bi-
pyramid : Second Order.

FIG. 235. — Hexagonal
Prism First Order.

«...

;

-"'-

.--''

;.:--

:.-.:*

.{TOT.

;•--;-;-

-.*-

«£L

L^*1

FIG. 2
Prism

36. — Hexagonal
Second Order.

angles (ab c). Any one of these may be taken as the vertical axis (c),
the other two being then lateral axes. The longer lateral axis is
called the macro-axis, the shorter is the brachy-axis.

XXIII

ATOMIC HEATS AND ISOMORPHISM

439

Pyramid (Fig. 237) and prism (Fig. 238) forms exist, but new
types of faces, known as domes and pinakoids, are met with in the
rhombic system.
Prism faces de-
veloped parallel
to one of the
lateral axes, and
intersecting the
other two axes,
are called dome
faces. If they
are parallel to
the longer, or
macro-axis, these

are called macrodomes (Fig. 239) ; if parallel to the shorter, or
brachyaxis, they are called brachydomes (Fig. 240).

Prism faces intersecting one lateral axis and parallel to the other

FIG. 237.— Rectangular Rhombic
Bipyramid.

FIG. 238.— Rhombic
Prism.

FIG. 239.— Dome and Pinakoid
Faces : Macrodome.

FIG. 240. — Dome and Pinakoid Faces :
Brachydome.

two axes are called pinakoid faces ; macropinakoids intersect the
macro-axis ; brachypinakoids the brachy-axis. These are the

101

^^^$^S^p5«

101

FIG. 241.— Barytes Crystal.

the faces 101 are the macrodome form,
the prism form are marked 00.1.

diamond-shaped end
faces in Figs. 239 and
240. In Fig. 241, re-
presenting a crystal of
barytes (BaS04), the
faces marked 010 con-
stitute a macropina-
koid form, or in this
case a basal pinakoid ;
The faces belonging to

440

INORGANIC CHEMISTRY

Sulphur occurs in pyramid forms of the rhombic system ; pyramid
and prism forms are shown by zinc sulphate, and stibnite (Sb2S3) ;
dome forms occur in aragonite (CaCO3), barytes, and potassium sulphate ;
pinakoid forms occur on crystals of anhydride (CaSO4).

(5) In the monoclinic system there are three axes, all of different
lengths ; two of the axes intersect one another at an oblique angle,
whilst the third is at right angles to the plane of the other two. Pyramid
and prism forms, pinakoids, and domes occur. The vertical axis is
denoted by c ; the 6-axis, or ortho-axis, is at right angles to the
vertical axis, whilst the inclined, or a-axis, is the clino-axis. The
angle between the vertical axis and clino-axis is called the angle /3.

FIG. 242.

Triclinic Crystal

(Copper Sulphate,

CuS04,5H2O).

FIG. 243.— Triclinic
Crystal (Potassium
Bichromate).

FIG. 244.— Relation of Tetrahedron
(Hemihedral Form) to Octahedron
(Hdlohedral Form).

An example of a monoclinic crystal is gypsum (Fig. 216) ; green vitriol,
washing-soda, borax, cane -sugar, and oxalic acid crystallise in this system.

(6) In the triclinic system there are three unequal axes intersecting
one another obliquely. One of these is selected as the vertical axis,
the other two are then spoken of as the macro-axis (longer), and the
brachy-axis (shorter). The three angles between the axes are also
given (a, /3, y). Examples of crystals belonging to this system are
copper sulphate (Fig. 242), potassium dichromate (Fig. 243), and
soda-felspar.

Hemihedral forms. — Those forms in any system which exhibit
the full number of faces required by the symmetry of the system are
called holohedral forms. If only half the number of faces occurring
in the holohedral form are present, the form is known as hemihedral.
Forms exhibiting only one quarter the full number of faces required
by the symmetry of the system are called tetartohedral. (In modern
classification into symmetry groups, these forms go as holohedral
forms into separate classes.)

Thus, a hemihedral form is produced by suppressing half the faces
of the holohedral form, and producing the remainder so as to meet
in new edges. Fig. 244 shows the form obtained by producing

XXIII

ATOMIC HEATS AND ISOMORPHISM

441

alternate faces of the regular octahedron : this is the regular tetra-
hedron, having four faces instead of eight. The tetrahedron is the
hemihedral form of the octahedron.

Important hemihedral forms occur in the hexagonal system. By
developing alternate faces of the hexagonal pyramid (Fig. 245), one

Fia. 245. — Hexagonal
Pyramid : Shaded
Faces to be Sup-
pressed.

FIG. 246.— Hemihedral
Form of Hexagonal
Pyramid : Positive
Rhombohedron.

FIG. 247.— Hemihedral
Form of Hexagonal
Pyramid : Negative
Rhombohedron.

obtains the positive or negative rhombohedron (Figs. 246, 247). From the
dihexagonal pyramid, with 24 faces, obtained by the combination of
two hexagonal pyramids, two kinds of hemihedral forms are produced':
(i) by suppressing alternate pairs of faces (Fig. 248) one obtains the
scalenohedron (Fig. 249) ; (ii) by suppressing alternate faces (Fig. 250)

FIG. 248.— Dihexagonal FIG. 249.— Scaleno- FIG. 250.— Dihexagonal FIG. 251. — Trapeze-
Pyramid : Alternate hedron: Hemihedral Pyramid : Alternate hedron : Hemihedral
Pairs of Faces to be Form obtained Faces to be Sup- Form obtained from
Suppressed. from Fig. 248. pressed. Fig. 250.

the trapezohedron (Fig. 251) results. Calcite (CaCO3) occurs as scalen-
ohedra and rhombohedra ; quartz occurs in trapezohedra : haematite
(Fe2O3), calamine (ZnCO3,), potassium and sodium nitrates, magnesite
(MgCO3), witherite (BaCO3), and strontianite (SrCO3) occur as
rhombohedra.

442

INORGANIC CHEMISTRY

CHAP.

Twin crystals. — Two or more individual crystals sometimes grow
in contact so that neither is complete, and twin crystals (Figs. 252 and

253) are formed.
The two crystals
may coalesce except
for a few faces, as
in Fig. 254.

Further particulars
of crystallographic
notation, etc., must
be obtained from the
regular text - books,
e.g., Tutton's " Crys-

FIQ. 253.— Twin Crystal of tallography " (Mac-
Gypsum, mfllan).
In the study of crystallography, however, the use of models, and the
examination of actual crystals, must accompany the reading.

FIG. 252. — Twin Crystal of
Fluorspar.

ISOMORPHISM

Isomorphism. — Haiiy (1743-1822), the founder of the science of
crystallography, laid down as fundamental axioms that : (i) iden-
tity of crystalline form (except in the cubic system) implies
identity of chemical composition ; and, conversely, (ii) difference
in crystalline form implies differ-
ence in chemical composition.

Exceptions to these statements
were, however, known at the end of
the eighteenth century. Klaproth
(1788) showed that calcium car-
bonate crystallised in the hexa-
gonal form as calcite, and in the
rhombic form as aragonite. Rome
de Flsle (1772) observed that,
from mixed solutions, copper
sulphate and ferrous sulphate
crystallise in the form of the

Toff™ TV«o alnmc aleo havp
latter. liie alums also nave

the same crystalline form, but
differ in chemical composition.

Mitscherlich (1820) cleared up these contradictions by show-
ing that phosphates and arsenates, when they were of
similar composition and contained the same amount of

FlG- 254.— Twins of Right- and Left-handed
yuartz. Partial and Complete Interpene-

tration.

xxm ATOMIC HEATS AND ISOMORPHISM 443

water of crystallisation, had almost exactly the same crystalline
form : e.g.,

Na2HPO4 -f 12H20, disodium hydrogen phosphate,
Na2HAsO4 -f- 12H2O, disodium hydrogen arsenate,

yield crystals of the same form. Haiiy's first axiom was therefore
disproved. In the case of the salts NaH2P04 -+- H2O and NaH2AsO4
+ H20 the ordinary crystalline forms were different, but the phos-
phate sometimes crystallised in a new form, identical with the
common form of the arsenate. Mitscherlich also discovered the
monoclinic variety of sulphur, showing that elements may have
different crystalline forms. Thus, one substance may assume two
distinct crystalline forms, and is then called dimorphous. If it
assumes more than two forms it is called polymorphous. Haiiy's
second axiom was, therefore, disproved. .

The capacity of different, but chemically similar, substances of
crystallising in the same form was called isomorphism by Mitscherlich ;
substances crystallising in the same form are called isomorphous.
Since, however, numerous analogous compounds of phosphorus and
arsenic, for example, are isomorphous, the latter name came to be
applied to the elements, arsenic and phosphorus, themselves.
Isomorphous elements are those which form similarly crystallising
compounds with the same elements or radicals : they can replace
each other in their compounds without causing any essential altera-
tion in the crystalline form. It is not necessary that the free
elements shall have similar crystalline forms, although this is some-
times the case.

Mitscherlich at first considered that the same number of atoms
combined in the same manner produce the same crystalline form, so
that the latter is independent of the chemical nature of the atoms,
and is deter mined solely by their number and mode of combination.
This generalisation proved to be too wide, and it was found that an
atom can be replaced by another without producing a change of form
only when the elements are chemically analogous.

More accurate measurements of crystal angles have shown, as
Mitscherlich conjectured, that the law is only approximate. Except
in the regular system, the replacement of an atom of one element
by an atom of an isomorphous element leads to a change in the
crystal angles which may, it is true, be small, but may amount to
several degrees. Haiiy's first axiom is, therefore, correct in the
strictest sense, although it is often only by refined measurements
that differences in the angles of crystals which are almost exactly
alike in appearance may be detected.

Thus, Tutton, in a long series of very exact measurements, found
that the crystal angles in isomorphous sulphates and selenates
of potassium, rubidium, and caesium, changed slightly when one

444 INORGANIC CHEMISTRY CHAP.

isomorphous element (K, Rb, Cs, or S, Se) was replaced by another.
The change, which may be expressed in terms of the ratios of the
lengths of the axes, a, 6, c, depends in a regular manner on the atomic
weight of the element.

K2SO4 a i b : c = 0-5727 : 1 : 0-7418 (K = 38-79)
Rb2SO4 a: b: c = 0-5723: 1 : 0-7485 (Rb = 84-77)
Cs2SO4 a: b : c = 0-5712 : 1 : 0-7531 (Cs = 131-6).

The other properties of the crystal (molecular volume =

^— : 2 — , refractive indices, coefficients of expansion, thermal

density

conductivity) altered with the crystal angles, showing that the
crystalline form is closely related to the nature of the atoms which
make up the structure of the crystal.

Isomorphous elements. — As a result of the investigations of crys-
talline form, it has been possible to classify the elements into eleven
groups, the members of each group being capable of replacing one
another without sensible alteration of crystalline form. The mem-
bers of each group are called isomorphous elements.

I. Cl, Br, I, F ; Mn (in permanganates, e.g., KMnO4 isomorphous with

KC104).

II. S, Se ; Te (in tellurides) ; Cr, Mn? Te (in the compounds K2RO4) ;

As and Sb in the glances MR2.

III. As, Sb, Bi ; Te (element) ; P, V (in salts) ; N, P (in organic

IV. K, Na, Cs, Rb, Li ; Tl, Ag.

V. Ca, Sr, Ba, Pb ; Fe, Zn, Mn, Mg ; Ni, Co, Cu ; Ce, La, Pr, Nd :

Er,Y with Ca ; Cu, Hg with Pb ; Cd, Be, In with Zn ; Tl with Pb.

VI. Al, Fe, Cr, Mn ; Ce, U in oxides R2O3.

VII. Cu, Ag in compounds of lower oxides ; Au.

VIII. Pt, Ir, Pd, Rh, Ru, Os ; Au, Fe, Ni ; Sn, Te.

IX. C, Si, Ti, Zr, Th, Sn ; Fe, Ti.

X. Ta, Nb.
XL Mo, W, Cr.

Several elements occur in more than one group. Thus, chromium
occurs in VI with Al, Fe, etc., because of the isomorphism of the

n in ii in

oxides R203, the spinels, R R'204, e.g., MgO, A12O3, or MgAl204,
FeO,Fe203, FeO,Cr203, etc., and the alums, e.g., K2S04, A12(SO4)3,
24H2O, K2SO4,Cr2(S04)3,24H2O, K2SO4,Fe2(SO4)3,24H2O. It
occurs in group II because of the isomorphism of the salts K2SO4,
K2CrO4, K2MnO4, etc. Manganese occurs in group V because of the

ii ii

isomorphism of the carbonates CaC03, FeC03, MnCO3 ; in group VI

xxiii ATOMIC HEATS AND ISOMORPHISM 445

because of the isomorphism of the spinels (containing Mn203, Fe2O3,
etc.) ; in group II because of the isomorphism of K2Mji04 with

VI VII

K2S04, etc. ; and in group I because of the isomorphism of KMnO4

VII

and KC104. The close connection between the valencies of an element
and its position in the groups of isomorphous elements is clear from
the above, and from a comparison of the table of isomorphous
elements with that of valencies (p. 252).

The formulae of similar compounds may be deduced from their
isomorphism : thus, from the fact that potassium selenate crystal-
lises in the same form as potassium sulphate, Mitscherlich
concluded that its formula must be K2SeO4, corresponding with
K2SO4. From its composition the atomic weight of selenium could
then be calculated.

Atomic weights from isomorphism. — The applications of isomor-
phism to the deduction of atomic weights are all based on the axiom
that isomorphous compounds have similar formulae.

Thus, ferric oxide, chromic oxide, and alumina are isomorphous,
since mineral crystals of these compounds have the same
form. The vapour density of aluminium chloride can be
found, and corresponds with the formula A1C13. The formula of
alumina will then be A1203. We therefore assume the formulas
Fe203 for ferric oxide and Cr2O3 for chromic oxide, and from the
percentage compositions of these oxides the atomic weights of the
metals may then be calculated. These are confirmed by the specific
heats of the metals, which are 0*1146 and 0-104, respectively.

The best example of isomorphism is probably that studied by Roscoe
in connection with the atomic weight of vanadium. The following
minerals had the formulae given assigned to them by Berzelius :

Apatite, 3Ca3(PO4)2 + CaF2, or Ca5P3O12F

Pyromorphite, 3Pb3(PO4)2 + PbCl2, or Pb6P3O12Cl

Mimetite, 3Pb3(AsO4)2 + PbCl2, or Pb5As3O12Cl

Vanadinite, 3Pb3V2O6 + PbCl2, or Pb5V3O9Cl.

In these formulae, lead and calcium, and arsenic and phosphorus,
replace each other, but the formula of vanadinite is quite different from
those of the other compounds, although all the minerals crystallise
in the same form. Roscoe therefore concluded either that the law of
isomorphism was incorrect or that Berzelius was in error in attributing
the above formula to vanadinite.

By reinvestigating the vanadium compounds, Roscoe was able to show
that the substance regarded as metallic vanadium by Berzelius was

446 INORGANIC CHEMISTRY CHAP.

really an oxide, VO. The formulae of the minerals were now completely
analogous :

Apatite, 3Ca3(PO4)2 + CaF2, or Ca6P3O12F

Pyromorphite, 3Pb3(PO4)2 + PbCl2, or Pb6P3O12Cl
Mimetite, 3Pb3(AsO4)2 + PbCl2, or Pb5As3O12Cl

Vanadinite, 3Pb3(VO4)2"-f PbCl2, or Pb6VsO12Cl.
The atomic weight of vanadium found by Berzelius, 68-5, was, there-
fore, in reality the molecular weight of the oxide VO, and the true value
was b8-5 — 16 = 52-5. Roscoe then found that the vanadium com-
pounds investigated by Berzelius contained phosphoric acid, which is
exceedingly difficult to separate. By using pure compounds he found
V = 51-4.

Formulas of minerals. — Since a certain amount of one element can
be replaced in a compound by an equivalent amount of an isomor-
phous element, the formula of the compound calculated from its
analysis will not usually give a whole number of atoms of each
isomorphous element.

Thus, spathic iron ore, FeCO3, may have the iron partly or com-
pletely (MnCO3) replaced by the isomorphous element manganese. The
relative proportions of the two metals may vary from Fe = 48-2 per
cent, and Mn = 0, to Fe = 0 and Mn = 47-8 per cent.

Such an isomorphous mixture is represented by a formula such as
(Fe,Mn)C03, the isomorphous elements being enclosed in brackets
with a comma separating them, and behaving as an equivalent
amount of one element. If follows from the law of isomorphism
that the sum of the atomic proportions of Fe and Mn, combined
with the group C03, must always be equal to unity.

Mixed crystals. — A very important property of isomorphous
substances is their capacity of crystallising together from solutions
so as to form homogeneous crystals containing the isomorphous
substances in variable proportions. Although the crystals are
perfectly homogeneous, they are usually known as mixed crystals ;
a more appropriate name is solid solutions. These are also formed
when the substances separate from a fused state.

Thus, if chrome alum, K2SO4,O2(SO4)3,24H2O, and ordinary potash
alum, K2SO4,A12(SO4)3,24H2O, which form deep purple and colourless
octahedral crystals, respectively, are dissolved together in water and the
solution is allowed to crystallise, octahedral crystals containing both
alums separate, having colours varying from a very pale purple to
deep purple according to the amount of chrome alum they contain,

Isomorphous compounds cannot, therefore, be separated in a
state of purity by crystallisation, as is the case with salts of dif-
ferent chemical types, crystallising in different forms, such as

xxm ATOMIC HEATS AND ISOMORPHISM 447

potassium nitrate and sodium chloride (p. 564). It is found that
substances which crystallise in the same form, but belong to different
chemical types, do not form mixed crystals, or only to a very
limited extent, whereas chemically analogous compounds may form
mixed crystals even though the* crystal angles differ by as much as
5° : the "resulting crystals have angles which lie between those of
the components.

Retgers (1889) considers the property of forming mixed crystals
to be a very important criterion of isomorphism. He also considers
that the variation in the physical properties of the mixed crystal
with the proportion of its constituents is a valuable guide in deciding
whether the substances are truly isomorphous or not. One of these
physical properties is the specific volume, i.e., the reciprocal of the
density, or the volume in c.c. of 1 gram of the substance. If this is
plotted against the proportions of the constituents, the points must,
according to Retgers, lie on a straight line which shows no change of
direction anywhere. The substances may be only partially miscible,
in which case there is a gap in the line, but if the substances are
isomorphous one part of the line is a continuation of the other, and
there is no change of direction.

Overgrowth crystals. — If an octahedral crystal of chrome alum is
suspended by a thread in a saturated solution of potash alum, a
colourless overgrowth of the latter salt is deposited on the violet
crystal as a nucleus. In the same way, a green crystal of nickel
sulphate, NiS04,7H20, may be covered with colourless zinc sulphate,
ZnS04,7H20. H. Kopp (1879) regarded the property of forming
overgrowth crystals as characteristic of isomorphous substances,
but exceptions to this criterion are known ; thus, rhombic K2S04
(pseudo-hexagonal) may form an overgrowth of hexagonal
KNaS04.

Exceptions to the law of isomorphism. — Exceptions to the law of
isomorphism are frequent. In some cases this may be due to the
existence of two or more varieties of a substance — dimorphism, or
polymorphism, respectively, only one of which, not the common
form, is isomorphous with the commonly occurring variety of a
chemically similar substance. An example of this was discovered
by Mitscherlich, viz., the acid phosphate and the acid arsenate of
sodium: NaH2P04,H20 and NaH2As04,H2O (p. 443). In many
cases, however, isomorphism is observed with substances exhibiting
no chemical similarities. Thus, the ammonium salts, containing
the radical NH4, are isomorphous with potassium and sodium salts
containing the atoms K and Na ; silver sulphide, Ag2S, in the mineral
argentite is isomorphous with lead sulphide in galena, PbS, the two
forming mixed crystals ; calcium carbonate, CaCO3, occurs in the
same form (aragonite) as sodium nitrate, NaN03 ; the compounds
Mg2Si04 and Al2Be04 are isomorphous ; and so on.

448 INORGANIC CHEMISTRY CHAP.

Other examples of isomorphism without similarity in chemical
composition are shown in the following groups :

(1) Potassium periodate, KIO4 (3) Potassium perchlorate, KC1O4
Calcium tungstate, CaWO4 » Barium sulphate, BaSO4
Potassium osmiamate, KOsO3N Potassium borofluoride, KBF4

(2) Potassium sulphate, K2SO4 (4) Yttrium phosphate, YPO4
Potassium beryllium fluoride, Zircon, ZrSiO4

K2BeF4 Tinstone, SnO2, or SnSnO4.

In these groups the molecule contains the same number of atoms,
and the original idea of Mitscherlich, that the form depended on the
number of atoms in the molecule, and not on their chemical nature,
is verified.

These exceptions to the law of isomorphism, although of great interest
to crystallography, have little significance in the chemical application of
the law ; the latter has served its purpose, and has led to the correction
of some atomic weights which have been confirmed by more certain
methods. There remains no possibility that these numbers will ever
be modified as a result of further compilation of lists of exceptions
to the law of isomorphism, and the latter have no further interest to
chemists.

EXERCISES ON CHAPTER XXIII

1. What methods are used in deciding which multiple of the equivalent
is the atomic weight of an element ? The composition of potassium
selenate, as determined by Mitscherlich, is : potassium 35-29 ; selenium
35-75 ; oxygen 28-96. The salt is isomorphous with potassium sul-
phate, K2SO4. Find the atomic weight of selenium.

2. The vapour density of aluminium bromide is 268 (H = 1). Its
percentage composition is: aluminium = 10-15; bromine = 89-85.
The specific heat of aluminium is 0-225. Find the atomic weight of
aluminium.

3. State Dulong and Petit's law. With what degree of exactness
does it hold, and what modifications of it have been proposed ?

4. What is known as to the atomic heats of elements at low tempera-
tures ?

5. What relation exists between the molecular heat of a solid com-
pound and the atomic heats of its constituents ? The specific heat of
nickel sulphide (NiS) is 0-1281. The specific heat of nickel is 0-1092 ;
find the specific heat of sulphur.

6. The specific heat of anhydrous calcium chloride is 0-1675, that of
the hexahydrate is 0'3461. Find the specific heat of the water of
crystallisation, and compare the number with the specific heat of ice
(p. 201).

7. What do you understand by a " crystal " ? What elements of
symmetry are met with in crystals ?

xxm ATOMIC HEATS AND ISOMORPHISM 440

8. What are the crystal systems ? To what systems do the following
belong : rock salt, alum, potassium dichromate, blue vitriol, calcite,
borax ?

9. Explain the meaning of : axis of symmetry, pinakoid, hemihedral
form, macrodome, twin crystal, dimorphism, isomorphous element.

10. What are the criteria of isomorphism ? What exceptions to the
law of isomorphism are met with ?

11. A mineral has the following percentage composition: Fe = 28 '6 ;
Mn = 19-5 ; CO, = 51'9. Find its formula.

G a

CHAPTER XXIV

THE CLASSIFICATION OF THE ELEMENTS AND THE PERIODIC LAW

Classification of the elements. — In classification, things are grouped
according to similarity ; those which resemble one another in some
respects being placed together, and those which are dissimilar being
separated.

Various criteria of likeness may be adopted, and it frequently
happens that things grouped according to one kind of likeness are
separated on the basis of another. The best classification will be
that in which the things grouped together in it resemble one another
in the greatest possible number of respects, each of which might
itself serve as the basis of a separate, and more specific, classification.

The classification of the elements into metals and non-metals
is one obvious^basis, although it presents certain difficulties. The
differences between metals and non-metals are in fact not always
sharply defined : the following are usually accepted as the most
important : —

METAJLS are electropositive elements (cf. p. 252) ; they normally give
basic oxides [acidic oxides are only formed when the atom has higher

II . VI1^

valencies, e.g., Mn=O (basic) ; K— O— Mn~O (acidic)] ; they

form halogen compounds stable in presence of water (KC1, PbCl2), or
decomposed only to a limited extent (BiCl3, SbCl3 ), the reaction being
reversible, BiCl3 + H2O ;=± BiOCl -f 2HC1 ; they form complex salts,
in which the metal may be present either in the electropositive radical
(cation), as in [Ag(NH3)2]Cl, or in the electronegative radical (anion),
as in K[AgC2N2].

NON-METALS are either electronegative elements, or show only very
feeble electrochemical properties (e.g., carbon) ; they give acidic oxides
in which the element has its normal valency (in some cases metallic
oxides with normal valency can function as feebly acidic oxides in the
presence of a strong base ; e.g., zinc oxide, ZnO, can give a stable
chloride, ZnCl2, with hydrochloric acid, or an unstable zincate, Zn(ONa)2,

450

CH. xxiv CLASSIFICATION OF ELEMENTS, PERIODIC LAW 451

with caustic soda) ; their halogen compounds are almost completely
decomposed by water : PC13 + 3H2O = H3PO3 -f 3HC1.

Certain physical properties commonly supposed to be characteristic
of metals are not so in reality :

lustre : this is shown by the non-metals iodine and carbon (graphite) ;

malleability : some metals (e.g., Bi, Sb) are brittle ; plastic sulphur
may be regarded as a malleable non-metal ;

high density : the alkali-metals are lighter than water (e.g., Li, sp. gr.
0-53), iodine has a density of 4-9 ;

conductivity for heat and electricity : graphite is a good conductor of
electricity, whilst some metals are relatively poor conductors, e.g.,
bismuth.

The classification of elements according to valency (cf. p. 252) is
not entirely satisfactory for two reasons : (1) the valen-cy of some
elements is variable ; (2) elements having the same valency often
differ in nearly every other respect (e.g., sodium is a strongly electro-
positive metal ; chlorine is a strongly electronegative non-metal ;
both are univalent elements).

The most satisfactory system of classification, and the one now
adopted, is based on the relation between the properties of the
elements and their atomic weights.

As early as 1817 Dobereiner noticed regularities in the atomic
weights of elements which were chemically analogous. In groups of
three such elements, the atomic weight of the middle element is
approximately the mean of the atomic weights of the extreme
elements. This is known as the law of triads. Thus in the following
groups (H = 1) this is observed :

/Cl 35-18 /S31-81

SO-54< Br 79-29; 79-16< Se 78-6 ;

M 125-91 \Tel26-5

/Ca 3-9-75

88-01< Sr 86-93

\Ba 136-28

Similar regularities were pointed out by Dumas, Pettenkofer,
Odling, and other chemists, but little additional progress was made
until a uniform set of atomic weights had been derived by Cannizzaro
(1858) from Avogadro's law, and the law of atomic heats. So long
as " equivalents," deduced from various considerations which had
no real bearing on the matter, continued to be used by different
chemists, no regularities could ever come to iight.

Newlands published a series of papers in the Chemical News,
beginning in 1863, in which he observed that if the new atomic weights
are used, and if the elements are arranged in the order of atomic
weights, " the eighth element, starting from a given one, is a kind of

Q G 2

452 INORGANIC CHEMISTRY CHAP.

repetition of the first, like the eighth note in an octave of music."
He called this the law of octaves.

1H 2 Li 3 Be 4B 5C 6 N 7O

8 F 9 Na 10 Mg 11 Al 12 Si 13 P 14 S

15 Cl 16 K 17 Ca 18 Cr 19 Ti 20 Mn 21 Fe, etc.

This relationship, which is based on what is now called the atomic
numbers of the elements, was not wholly satisfactory, as can be seen.

MENDELEEFF.

It was received with coolness by the London Chemical Society,
which declined to publish Newlands's paper in its journal.

Although the germ of one of the most important chemical laws is
contained in this table, the credit of having stated clearty the con-
nection between the properties of the elements and their atomic
weights, of forcing this result on the attention of contemporary

xxiv CLASSIFICATION OF ELEMENTS, PERIODIC LAW 453

chemists, and of making it the foundation of a comprehensive
system of classification, belongs without question to the great
Russian chemist, Dmitrij Ivanovitsch Mendeleeff (1834-1907), born
at Tobolsk in Siberia, and professor at St. Petersburg from 1866 to
1890. Whilst engaged in writing his classical " Principles of
Chemistry " (1st edition, 1869 ; 3rd English edition, 1905) Mendeleeff
had ample opportunity of studying the properties of the elements,
and during that period an idea arose in his mind, which, unlike the
majority of ideas occurring to investigators, turned out to be a
fundamental law.

The Periodic Law.— -The basic idea of Mendeleeff 's system of
classification was that " there must be some bond of union between
mass and the chemical elements ; and as the mass of a substance is
ultimately expressed in the atom, a fundamental dependence should
exist and be discoverable between the individual properties of the
elements and their atomic weights. But nothing, from mushrooms to
scientific dependence, can be discovered without looking and trying.
So I began to look about and write down the elements with their
atomic weights and typical properties, analogous elements, and like
atomic weights on separate cards, and this soon convinced me that
the properties of the elements are in periodic dependence upon their
atomic weights ; and although I have had my doubts about some
obscure points, yet I have never once doubtgd the universality of
this law, because it could not possibly be the result of chance."
(" Principles of Chemistry," II, p. 30, 1905.) Immediately after
the publication of this Periodic Law by Mendeleeff in 1869, an
identical generalisation was put forward independently by Lothar
Meyer in 1870.

Mendeleeff from the first was convinced of the accuracy of the
law, and did not hesitate to alter some of the accepted atomic
weights on that ground : Lothar Meyer was doubtful, believing that
" it would be rash to change the accepted atomic weights on the
basis of so uncertain a starting point." Further work has, with
one doubtful exception (tellurium), confirmed the changes boldly
advocated by the Russian chemist.

The essence of the Periodic Law is contained in Mendeleeff 's state-
ment, quoted above. As the atomic weights progressively increase,
the properties of the elements alternately ebb and flow. The
heights of the tide, the alternation of day and night, and of the
seasons, are in the same way in periodic dependence on the uniform
march of time, the fundamental independent variable.

Atomic volumes. — In testing the Periodic Law it is desirable to
use such properties of the elements as can be expressed numerically.
One of these is the atomic volume, i.e., the volume in c.c. of the atomic

454 INORGANIC CHEMISTRY CHAP.

weight in grams of a solid element, or iri other words the atomic
weight divided by the density :

Atomic weight A
Atomic volume = - =

These atomic volumes represent, not the space occupied by the
atoms themselves, but this plus the empty spaces between the
atoms in the material. If the atoms are assumed to be spherical,
**J A]D is a measure of the mean distance between the atomic
centres.

The atomic volumes of a few important elements are given below.

Element. fc^ Density at 16'

Hydrogen ...... 1 0-086at-250° 11-6

Lithium ... ... 6-9 0-534 12-9

Sodium ...... 22-8 O971 23*5

Potassium ... 38-8 0-862 45-0

Rubidium ...... 84-8 1-532 55-2

Caesium ...... 131-8 1-87 70-5

Calcium ...... 39-8 1-55 25-7

Strontium ... 87-0 2-54 34-3

Barium ...... 136-3 3-75 36-4

Chlorine ...... 35-2 2-49 (liq. at 0°) 14-1

Bromine .'.. ... 79-3 3-102(25°) 25-5

Iodine ...... 126-0 4-95 25-5

Iron ...... 55-4 7-86 7-05

Lead ...... 205-5 11-37 18-1

Mendeleeff remarked that comparatively light and reactive
elements have the largest atomic volumes (Na, K, Rb, Cs, halogens) ;
elements which are not very reactive have small atomic volumes
(C as diamond, Ni, Co, Ir, Pt). Lothar Meyer plotted the atomic
volumes against the atomic weights, and obtained the atomic
volume curve shown in Fig. 255. This curve also exhibits periodicity
in the case of other properties, such as expansion by heat, conduc-
tivity for heat and electricity, magnetic susceptibility, melting
point, refractive index, boiling point, crystalline form, compressi-
bility, atomic heat (plotted as the thin curve in Fig. 255), heats of
formation of oxides and chlorides, hardness, malleability, volatility,
volume change on fusion, viscosity and colour of salts in aqueous
solution, mobilities of ions (p. 288), electrode potentials of
metals (p. 884), over- voltage of metals, frequency of atomic vibra-
tions in solids (p. 430), distribution of the elements in nature
(p. 32), distribution of lines in spectra (p. 756), and valency.

xxiv CLASSIFICATION OF ELEMENTS, PERIODIC LAW 455

As Mendeleeff said, " these regularities can hardly be the result
of chance."

Lothar Meyer pointed out that ductile metals of low density occupy
the maxima (Li, Na, K, Rb, Cs), or descending parts of the curve
near the maxima (Mg, Ca, Sr, Ba). Ductile metals of high density
occupy the minima (Al, Fe, Ru, Ce, Ir), and adjacent parts of the ascend-
ing curves (Ni, Cu, Pd, Ag, Pt, Au). Brittle metals of high density
occur on the descending parts of the curve shortly before the minima
(Ti, V, Cr, Mn, Sb, Bi : exceptions are Ta, W, Ir.)

Fusibility and volatility. — The fusibility and volatility of elements,
given by their melting and boiling points, are also in periodic
dependence on the atomic weights. All the gaseous elements and
those fusing readily below a red heat (see table), occur at the
maxima and on ascending portions of the atomic volume curve.
Difficultly fusible elements occur at the minima or on descending
portions of the curve.

Carnelly found a similar periodic dependence of the melting points
of the metallic chlorides, and the heats of formation of the oxides
and chlorides, on the atomic weight of the metal : the periodicity
of properties thus extends to the compounds of elements.

Electrochemical character. — The electrochemical character of an
element, which will be more closely considered later (p. 886), is
roughly denned by the chemical character of its oxide : electropositive
elements yield basic oxides (p. 450), whilst electronegative elements yield
acidic oxides. If the part of the atomic volume curve situated
between two maxima is called a section, then all elements on descend-
ing parts of the second and third sections are electropositive ;
those on ascending portions are electronegative. Elements situated
on sections 4 and 5 exhibit electrochemical properties passing
through two periods whilst the atomic volumes pass through oijly
one. On the first portion of the descending curve of each of these
sections, strongly electropositive elements occur (K, Ca ; Rb, Sr) ;
these are followed, on the same part of the curve, by more or less
electronegative elements (V, Cr, Mn ; Zr, Nb, Mo, Ru, Rh), which are
again followed, on the ascending portions of the curve, by electro-
positive elements (Fe, Ni, Co, Cu, Zn, Ga ; Pd, Ag, Cd, In) ; finally,
after these on the same but higher parts of the curve, come electro-
negative elements (As, Se, Br ; Sn, Sb, Te, I). The sixth and seventh
sections are considerably broken up, but similar regularities are
noticed.

Atomic heats, — The atomic heats of solid elements are, at the
ordinary temperature, practically constant and equal to 6-3.
The curve representing them on the atomic weight diagram is
therefore a horizontal straight line. It was formerly con-

456

INORGANIC CHEMISTRY

CHAP.

sidered that the atomic (or the specific) heat was an ex-
ceptional property with respect to periodicity: the more accu-
rate investigations of Dewar at low temperatures (1913) showed,
however, that the atomic heats at low temperatures, when plotted
in terms of the atomic weights, reveal definitely a periodic variation
closely resembling the atomic volume curve. The similarity of the
two curves suggests that, at low temperatures, equal volumes of
different elements, instead of equal numbers of atoms, have the same
capacity for heat (Fig. 255).

Compressibilities. — Since the compressibilities of solid elements are

10 20 30 4O 5O CO 70 80 90 1OO HO 120

Atomic Weights

Fia. 255.— Dewar's Atomic Heat Curve with a Curve of Atomic Volumes.

in periodic dependence on the atomic weights, T. W. Richards
supposes that this indicates that the atoms themselves are
compressible. It is very improbable that such a conclusion is
justified in the sense understood by Richards. It is more likely
that the atomic forces, which resist the approach of atoms situated
at small distances from each other, are dependent on the masses
of the atoms, or their electrical constitution, in a periodic
manner.

The periodic law.— The original statement of Mendeleeff (1869)
includes practically the whole content of the Periodic Law. It
is given in eight paragraphs :

xxiv CLASSIFICATION OF ELEMENTS, PERIODIC LAW 457

" (1) The elements, if arranged according to their atomic weights,
exhibit an evident periodicity of properties.

" (2) Elements which are similar as regards their chemical pro-
perties have atomic weights which are either of nearly the same
value (platinum, iridium, osmium), or which increase regularly
(potassium, rubidium, caesium).

" t3) The arrangement of the elements, or of groups of elements, in
the order of their atomic weights corresponds with their so-called
valencies.

" (4) The elements which are the most widely distributed in nature

Atomic Volume

130

140

.160

170

(From Lowry : " Historical Introduction to Chemistry," Macmillan.)

have small atomic weights, and . . . sharply defined properties.
They are therefore typical elements.

" (5) The magnitude of the atomic weight determines the character
of an element [and those of its compounds.]

" (6) The discovery of many yet unknown elements may be ex-
pected, for instance elements analogous to aluminium and silicon,
whose atomic weights would be between 65 and 75. [These have
since been discovered.]

" (7) The atomic weight of an element may sometimes be corrected
by the aid of a knowledge of those of the adjacent elements. [This
has been done in several cases.]

458

INORGANIC CHEMISTRY

CO I 1O Tt< O I I

CO I "* 1C <N I
i (N <M (M <N

' O O O

• <N O I O

iO »O <N | t-

<M CO <N <M CO CO

° *.

<D ^_j OT^Tt(TtH-^iiOiOCDI>C5T^t~-I>OOOOO5^i,CO'^lOt^

2^ "^frS • A

•. * (M

« S : : : : S S "

.2 S -.2-2.3 s a .5 y 9 S S fi's-SS^^d

LJ rt ^-t -ti i-* * FT^ d rH 3 3 ""H ^«S P.^^ — i H (1s! rt r^ *75 m ^
0)

a f llgs^Ss^gg^sg^ssiSia^sa
S ^ §

^ ! °.£S£§2i3 188 i -|- 1 | |2 | | ||lg

•^ i— l <J<> C^

•

"?•

§ _^ »

g g : : g d | ; : g R : «
e ^

CJ g

3 '^

PQ >o

^ »o

p, ' 9^ .T ^^coo «oi>oot-4 4 ooo

COlOCOOOOOOiOOlOOCOlO CD |cD(Mt>OOi-H -^1 OTt<CD
(NC<J<M>-Hi— I i— ii— i i— ii-HCO I-HI— i

O I I I I M M I I , — • — . A

l> Tt<

XOGOCO OCplOGOC<JQO i— I I

-Hi-Hi— I i-H i— I i— I r-( >— I FH t—

i i i i i i S rH

rH rt S <M

|l:|g|a:!c.ir|||li|soi I |ll

H I ill

xxiv CLASSIFICATION OF ELEMENTS, PERIODIC LAW

459

" (8) Certain characteristic properties of the elements can be fore-
told from their atomic weights." [<?/. (6).]

The periodic system : Mendeleeff arranged the elements in a table,
called the periodic table, or periodic system, an abbreviated form of
which is given below. In this the elements are arranged in
nine vertical columns, called groups, headed by zero (0) and
the Roman numerals from I to VIII. These groups arise
by suitably breaking up, into assemblages, a continuous series
of the elements arranged in the order of their atomic weights.
These assemblages are called periods, and if the periods are written
one beneath the other in horizontal rows, the elements of all the
periods which are vertically beneath each element in the first period
constitute the groups. The first complete period contains eight
elements, from helium to fluorine ; the next period also contains
eight elements, from neon to chlorine, and elements of the second
period are analogous to those vertically above them in the first
period. In other words, the periodicity of properties begins again
after fluorine, and the same types of properties are met with in the
ninth, tenth, eleventh, etc., elements, as in the first, second, third,
etc.

PERIODIC TABLE.

GROUP.

0. I. II. III. IV. V. VI. VII.

H -

VIII.

Fe Co Ni

Ru Rh Pd

Sii

15 Rare

Os Ir Pt

Earths

—

- •

After chlorine, however, eleven elements, instead of eight, must
be passed over before the periodic recurrence of properties begins

460 INORGANIC CHEMISTRY CHAP.

again. At the beginning of this period we also meet with the -first
serious difficulty in the periodic system, viz., that the element next
in atomic weight to chlorine is potassium, which undoubtedly
belongs to the same group as sodium. The next element is argon,
which is an inert gas resembling helium and neon, and therefore
belonging to the zero group. The order of the two elements in
respect of their atomic weights is therefore the inverse of the order in
the periodic system which brings them into the same groups as their
chemical analogues. In such cases, where the atomic weights are
apparently inverted, the elements are placed in the groups to which
they naturally belong, and the atomic weights disregarded. Three
such pairs of elements are known (H = 1) :

1. A 39-6 ; K 38-8. 2. Co 58-50 ; Ni 58-21.
3. Te 126-5; I 125-91.

With this transposition of argon and potassium, the natural
sequence runs along the period until manganese is reached. We
should then expect an inert element resembling argon. Actually we
encounter three elements, iron, cobalt, and nickel, with atomic
weights almost identical, and resembling one another very closely
in their physical and chemical properties. After these three elements
come copper, zinc, etc., which resemble in some respects the elements
of Groups I, II, etc., and the inactive element does not appear. The
three elements iron, cobalt, and nickel are placed in a separate
group, viz., Group VIII, no representatives of which exist in the
preceding periods, and the elements following, viz., copper, zinc, etc.,
which do not closely resemble the preceding elements of the same
groups, are separated from these by placing them on the right in the
group, whilst the other elements are placed on the left.

The three elements in Group VIII belonging to this period are
called transitional elements, and instead of two short periods of eight
elements in each, the whole of the 18 elements from argon to bromine,
inclusive, are regarded as forming a long period, divided into elements
belonging to even and odd series according as they occur in series of
even (e.g., Ca) or odd (e.g., Cu) number, beginning with hydrogen.

This first long period is followed, after bromine, by a second long
period, beginning with the inactive element krypton, followed by the
odd elements of the period as far as molybdenum. These elements
resemble the corresponding odd elements of the preceding long period.
But after molybdenum we should expect an element resembling
manganese. Instead of this, we find a cluster of three elements with
very similar atomic weights, and with closely allied chemical pro-
perties, viz., ruthenium, rhodium, and palladium. These obviously
are transitional elements, of the same type as iron, cobalt, and nickel,
and must therefore be placed in Group VIII. This leaves a gap in
Group VII, which we must assume should be occupied by a missing

xxiv CLASSIFICATION OF ELEMENTS, PERIODIC LAW 461

element, not yet discovered, which should resemble manganese.
The even elements of the long period then follow, ending with iodine.

At this point there is a repetition of the inversion of atomic
weights previously met with in the case of argon and potassium.
Iodine is undoubtedly a halogen element, belonging to Group VII,
whilst tellurium is equally certainly an element of Group VI, which
contains its chemical analogues, sulphur and selenium. In the
order of atomic weights, however, the positions would be reversed.
Again we disregard the atomic weights and place the two elements
in those positions which are in conformity with their chemical
•properties.

A new period begins with xenon, and proceeds as far as lanthanum,
in Group III, in a regular manner. After lanthanum, however,
comes a series of thirteen elements, with atomic weights differing
by one or two, or even four, units, all of which are most closely
analogous in chemical properties, and very difficult to separate in
analysis. These are the elements of the rare earths. Two elements
of the rare earths, viz., scandium and yttrium, occur in previous
periods ; it is obviously impossible to proceed in the normal
manner with the fourteen elements now encountered :

La Ce Pr Nd - Sa Eu Gd Tb Dy Ho Er

138-0 139-2 140-0 143-2 - 149-3 150-9 156-1 158-0 161-3 162-3 166-4

Tmi Tm2 Yb Lu

"TgfT' 172'2 173'7'

In this case, instead of one element occupying a place in the
group in the normal manner, we encounter a cluster of fourteen.
These must be placed in the same group as scandium and yttrium ;
the regular periodic change in the properties of the elements is
checked at this point, and goes forward again only when the atomic
weight has increased by about 40 units.

After the rare-earth elements the table becomes, apparently,
somewhat incomplete (see the full table on p. 466). From the
chemical properties of its quadrivalent compounds, cerium is
usually placed in Group IV. After cerium the next element is
tantalum, with an atomic weight about six units higher than
lutecium, the last element of the rare-earth group. ' It may
therefore be inferred, since the average difference of atomic
weights of successive elements in this period is about three units,
that an unknown element of the rare earths should come between
lutecium and tantalum, and occupy Group III, since cerium is given
the place in Group IV, and tantalum is, from its chemical character-
istics, obviously an element of Group V. After tantalum comes tung-
sten, then three transitional elements : osmium, iridium, and plat-
inum, which fall in Group VIII, leaving a blank for an unknown
element in Group VII. The rest of this long period is continued

462

INORGANIC CHEMISTRY

CHAP.

PERIODIC SYSTEM

200-

FIG. 256. — Arrangement of Elements in Periodic System
according to Harkins and Hall.

from gold, in Group I,
to bismuth, in Group
V. The final period
contains radioactive
elements. In this part
of the periodic table a
very remarkable cir-
cumstance has lately
come to light, viz., that
one place can be occu-
pied by a group of ele-
ments having very
nearly the same atomic
weights, and chemically
inseparable : these ele-
ments are called isotopes
(p. 114). Thus, the
position marked " Pb"
(lead) in the table is in
reality occupied by
ten elements, some of
which are products of
radioactive changes, as
will be explained later,
all chemically insepar-
able but having dif-
ferent atomic weights :

Atomic weight
(H = 1

1. Lead from

radium ... 204-5

2. Ordinary lead 205 -6

3. Lead from

thorium ... 206-5

4. Radium D 208-4

5. Thorium B 212-4

6. Radium B 210-4

7. Lead from
actinium ... ?

8. Lead from

actinium B ?

9. Product of
branch -chain,
radium series 208-4

10. Product of
branch-chain,
actinium series ?

xxiv CLASSIFICATION OF ELEMENTS, PERIODIC LAW 463

A similar state of affairs is found in the place occupied by bismuth,
which includes ordinary bismuth, radium C, radium E, actinium C,
and thorium C ; and also in other positions.

The close resemblance of elements occupying odd or even periods,
and their difference from elements of even or odd periods, respec-
tively, in the same group, has been mentioned. Thus, the odd
series in Group VI comprises Cr, Mo, W, U, and the even series
S, Se, Te. It is only elements taken from an odd or even series in a
group which obey the law of triads (e.g., Ca, Sr, Ba ; or Zn, Cd, Hg,
in Group II).

Several space-models of the periodic system have been devised :
the latest is that due to Harkins and Hall (1915), which is shown in
Fig. 256. It consists of a double spiral, and the relationships
between the elements are readily seen.

The periodicity of valency. — Mendeleeff in his original statement
of the Periodic Law (p. 457) pointed out that the number of the
group in the system corresponds with the valency of the elements
occurring in it. In some cases (e.g., N, S, Cr, Mn, Fe, etc.) it is-
necessary to attribute to an element its maximum valency, in
others (Cu, Ag, Au) .the minimum valency, and the assignment
of valency therefore seems a little artificial. This has been
urged as a defect of the periodic system by Wyruboff (1896),
but if we keep in mind how little is really known of valency,
the discrepancies are not serious. The valencies are most
clearly seen in the different groups when oxygen compounds are
considered :

VIII
(Ru208)

(Os2o8)

Group

III

VII

Li20

(Be202)

B203

(C204)

—

Na2O

(Mg202)

A1203

(Si204)

(SA)

CIA

K2O

(Ca202)

Se203

(Ti204)

(Cr206)

[Mn207]

Cu2O

(Zn202)

Ga203

(Ge204)

As2O5

(Se206)

[Br2O7?J

Rb2O

(Sr202) '

Y203

(Zr204)

Nb205

(Mo206)

—

Ag20

(Cd202)

In203

(Sn204)

Sb2O5

(Te206)

[I207]

Cs20

(Ba202)

La203

(Ce204)

Ta205

(W206)

—

Au20 (Hg202) T1203 (Pb204) Bi20

The inactive gases, which form no compounds and have therefore
zero valency, occupy the zero group. This group separates the
intensely electronegative elements of Group VII from the intensely
electropositive elements of Group I. The transitional elements
of Group VIII perform this function for the three parts of long
periods where there is no inactive element. In these cases, how-
ever, the negative and positive properties of the elements in the

464 INORGANIC CHEMISTRY CHAP.

first and seventh groups are much less marked than in the cases
where inactive elements are interposed :

F Ne Na; Cl A K; Mn (Fe, Co, Ni) Cu ; Br Kr Rb ;
Mo (Ru, Rh, Pd) Ag ; I Xe Cs ; W (Os, Ir, Pt) Au.

An important distinction between elements of the odd and even
series is the capacity of the former alone to form organo- metallic
compounds, i.e., compounds of metals (sometimes of non-metals
closely related to metals, e.g., boron) with hydrocarbon or other
similar radicals, e.g., NaCH3, Zn(C2H5)2, Pb(C2H5)4. These are
not formed by elements of the even series except of Group VIII.

The elements of Group VIII are distinguished by the facility
with which they form complex compounds, especially those con-
taining cyanogen or ammonia : potassium ferrocyanide, K4[FeCy6] ;
hexammine cobaltichloride, [Co(NH3)6]Cl3.

Electrochemical character. — The strongly electropositive elements
are associated towards the left of the table, beginning with Group I ;
the elements of strongly electronegative character occur on the
right of the table, the most marked being in Group VII. In passing
along the period from Group I to Group VII the electropositive
character diminishes. When Group IV is reached the elements
show hardly any electrochemical character, either positive or
negative, and are practically neutral. The electropositive character
then changes. over into electronegative, which becomes increasingly
stronger until it reaches a maximum in Group VII. The electro-
chemical character is well shown in the oxides of elements of the
second period :

Na2O MgO A12O3 SiO2 P2O5 SO3 C12O7

strongly basic weakly weakly fairly strongly very

basic basic and acidic strongly acidic strongly

acidic acidic acidic

The gradation of electrochemical character is shown also in the
groups themselves.

The last members of the even series resemble in many respects
the first members of the next odd series (excluding the zero group).
Thus, there is a gradual transition from chromium and manganese
to copper and zinc, apart from the bridge formed by the transitional
elements. This gradation of properties in the periods was insisted
upon by Mendeleeff, but has been somewhat neglected in com-
parison with the more obvious gradations in the groups.

Differences between atomic weights. — The differences between
the atomic weights of successive elements in the various periods
show many striking regularities, which have been the object of
interesting speculations. The earliest of these, dating back to

xxiv CLASSIFICATION OF ELEMENTS, PERIODIC LAW 465

1815, long before the discovery of the periodic relationship, is
Front's hypothesis. According to this, the atomic weights of 'the
elements are whole multiples of that of hydrogen. In 1816 Prout
stated that the simplest explanation of this supposed regularity
is to be found in the assumption that the atoms of all elements are
formed by the condensation of a greater or smaller number of
atoms of hydrogen, this element being the primary matter or protyle
(Greek prote, first ; hule, stuff, cf. p. 28).

Prout 's hypothesis, although disproved in its original form by
the accurate determinations of atomic weights made by Berzelius,
still had an extraordinary fascination for chemists. When Dumas
and Stas in 1841 redeter mined the atomic weight of carbon, finding
it almost exactly 12, and showing that Berzelius had made an error
in this case of no less than 2-5 per cent., the figures of the latter
were looked upon with great mistrust, which subsequent work has
not justified.- When, further, these two experimenters found that
the atomic weight of oxygen was almost exactly 16, interest in
Prout's hypothesis revived. The atomic weight of chlorine, how-
ever, was found to be nearly 35-5, so that Dumas suggested that
atomic weights are multiples of half the atomic weight of hydrogen.
Marignac (1860) suggested that the law of constant proportions
might not be quite exact, but that small variations of composition
might occur, which would explain the deviations from whole
numbers. Stas, beginning with " an almost complete confidence
in the exactness of the law of Prout," was led by his exact researches
to conclude that it "is only an illusion, a pure hypothesis definitely
contradicted by experiment." The hypothesis of Prout was also
very definitely rejected by Mendeleeff.

Interest in Prout's hypothesis revived as a result of the experi-
ments of Crookes (1887) on the discharge of electricity through
highly rarefied gases. As a result of this work, Crookes was led
to assume that the electricity is carried in vacuum tubes by a
" fourth state of matter," as much more attenuated than ordinary
gases as the latter are removed from the liquid state. This fourth
state of matter he identified with protyle, and regarded the atoms
of the elements as condensations of the primary matter. Later
investigations have confirmed these bold speculations in a surprising
manner (p, 1035).

Apart from hypothetical considerations, a number of interesting
regularities appear from an inspection of the periodic system
itself.

The mean difference between the atomic weights of correspond-
ing elements in series 2 and 3 is 16 ; that in series 3 and series 4
is 18J. The average difference between the atomic weights in
series 4 and 6 and 5 and 7 is 46, which persists between series 6
and 8. until the group of rare earth elements is reached, when it

H H

PERIODIC SYSTEM OF

SERIES

GROUP 0.

GROUP I.

GROUP II

GROUP III-

GROUP IV.

. —

R2O
RH

R202
RH?

R203

R204
RH4

1
2

He 3-97

H 1-000

Be 9-00

B10-8

C 11-910

Li 6-89

Ne 20-0

Na 22-82

Mg24-13

Al 26-8

Si 28-1

A 39-6

K 38-79

Ca 39-75

Sc 44-7

Ti 47-72

Cu 63-07

Zn 64-85

Ga 69-5

Ge-71-9

Kr 82-26

Rb 84-77

Sr 86-93

Yt 88-62

Zr 89-9

Ag 107-04

Cd 111-51

In 113-9

Sn 117-8

Xe 129-2 '

Cs 131-76

Ba 136-28

La 137-9 (and
12 other
elements of
RareEarths)

Ce 139-15

Aii 195-6

Hg 199-0

Tl 202-4

Pb 205-55

Ra-C2
Ac-D
Th-D

Pb ex Ra-C2
Pb ex Ra-F
Pb ex Ac-D
Pb ex Th-Ci
Pb ex Th-D
Ra-B
Ra-D
Ac-B
Th-B

Nt or Ra-
Eman
220-6

—

Ra 224-2

Ac-X

Ac?
MsTh2

Th 230-31

Ac-Eman
Th-Eman.

Ms Thi
Th-X

U-Y
Rd-Ac
Rd-Th

466

THE ELEMENTS.

GROUP V,

GROUP VI.

GROUP VII

R205
RH,

R206
RH2

R207
RH

N 13-897

O 15-87

F 18-9

P 30-79

S 31-81

Cl 35-18

GROUP VIII.

V 50-6
As 74-37

Cr 51-6

Se 78-6

Mn 54-49
Br 79-29

Fe 55-40 Co 58-50 Ni 58-21

Nb 92-4
Sb 119-2

Mo 95-2
Te 126-5

I 125-91

Ru 100-9 Rh 102-1 Pd 105-9

Ta 180-1

Bi 206-4

Ila-Ci
Ra-E
Ac-C
Th-C

W 182-5

Po or Ra-F
Ea-A
Ra-C
Ac-A
Th-A
Th-Ci

—

Os 189-4 Ir 191-6 Pt 193-6

•

EkaTa
U-X2

U-i 236-3
U-n

467

H H 2

468 INORGANIC CHEMISTRY CHAP.

rises to 89. The value falls again to 54 when the rare earths are

Rydberg (1914) observed that the two short periods, from Li to Ne,
and from Na to A, contain altogether 2 X 8 = 42 elements. The two
long periods, from K to Kr, and Rb to Xe, contain 2 X 1 8 = &' elements.
These should, if the same regularity holds good, be followed by two very
long periods containing 2 x 32 = 82 elements, of which 53 are known
from Cs to U. On the other hand, we should expect the two short
periods to be preceded by periods containing 22 elements, of which
helium and hydrogen are known. Helium may be considered as the
fourth element ; experiments on the scattering of X-rays by gases
indicate consecutive positions for hydrogen and helium, so that two
hypothetical gases should precede hydrogen. Rydberg, however,
considers that these elements should come between H and He, and
identifies them with coronium and nebulium, evidence of the existence
of which has been found in the spectra of nebulae. Nicholson, on the
other hand, from spectroscopic evidence, believes that the upper period
contains, besides hydrogen, the hypothetical elements protohyolrogen
(0-081), nebulium (1-31), protofluorine (or coronium, 2-1), arconium
(2-9), etc. The existence of coronium (supposed by Mendeleeff to have
an atomic weight 0-4) has been inferred from the bright green lines seen
in the spectrum of the sun during the eclipse of 1869 ; although traces
of it were said to exist in volcanic gases by Nasini, Anderlini, and
Salvador! (1893), its presence on the earth is doubtful. Mendeleeff
also regards the luminiferous ether as an inactive element of atomic
weight about 10~6.

Correction of atomic weights. — Mendeleeff found it necessary to
alter some atomic weights in use in 1869 in order that the elements
should fall into those positions in the periodic table assigned to
them by their chemical properties. Thus, indium, which occurs
with zinc in minerals, has an equivalent of 37 -9. From its occurrence
with zinc, the element was supposed to be bivalent, the oxide being
InO ; hence the atomic weight would be 37-9 X 2 == 75-8. The
element should then go in Group II after zinc, but this position
is occupied by strontium (87), and there is no place for an element
of atomic weight 754 in that group. There is also no place between
As = 75 and Se = 79, so that this atomic weight is impossible.
The vapour density, atomic heat, and isomorphism methods had
not been applied, so that there was no guidance available. Men-
deleeff pointed out that if indium is tervalent, its oxide being In203,
its atomic weight would be 37-9 X 3 = 113-7, when it would fill
a vacant space in Group III, between Cd = 112 and Sn = 118,
in the preceding and following groups, respectively. The chemical
and physical properties agree with this position. Thus, the densities
are Cd 8-6, In 74, Sn 7-2 ; the basic properties of In203 are inter-
mediate between CdO and Sn02 ; finally, the specific heat of
indium was found to be 0-055, indicating an atomic weight of
6-3/0-055 = 114-5. The element was then found to form alums,
and therefore belongs to Group III.

xxiv CLASSIFICATION OF ELEMENTS, PERIODIC LAW 469

Again, beryllium, with the equivalent 4-55, seemed to show
many resemblances to aluminium. The hydroxides of both
elements are gelatinous precipitates soluble in acids and
alkalies ; the carbonates cannot be prepared by precipitation, but
decompose immediately ; and the metals, obtained by the electrolysis
of the double fluorides of aluminium and beryllium with potassium,
dissolve in alkalies with liberation of hydrogen. The determination
of the specific heat of beryllium, finally, led to the value 13-65 for
the atomic weight. All these results appeared to point to beryllium
being tervalent, the oxide Be2O3 resembling A1203. But there is
no place for an element of this atomic weight in the first period :
Bll 012 N 14 O16. AvdeefT (1819) had previously pointed
out the analogy of the sulphate with that of magnesium, and
MendeleefT placed beryllium in Group II, before Mg, thus con-
sidering it to be bivalent, and its oxide BeO. Its atomic weight
should then be 4 -55 X 2 = 9 -1 , and there is a vacant place between
Li = 7 (univalent) and B = 11 (tervalent) for such a bivalent
element. Humpidge then found that the specific heat of beryllium
increases rapidly with the temperature, becoming 0-6206 at 500° :
this gives Be = 9-8. Nilson and Pettersson (1884) also found
that the vapour density of beryllium chloride was 40, which agrees
with BeCl2 (9 + 71 = 80), but not with BeCl3 (13-65 + 106-5 =
120-15). These chemists, therefore, abandoned their advocacy of
the tervalent character of beryllium.

In other cases the correction in the atomic weight amounted to
a few units only, the valency remaining unaltered.

Thus, gold was formerly placed before iridium, platinum, and osmium,
in the order given, in the atomic weight sequence (H = 1) :

Au 194-6 Ir 195-1 Ft 195-1 Os 197-0

Chemical analogies in the periodic system strongly suggest the order :

Os 189-4 Ir 191-6 Pt 193-6 Au 195-6

and more exact determinations gave the atomic weights shown.

Prediction of missing elements. — It has been mentioned that,
in arranging the elements in the periodic system, Mendeleeff had
to leave gaps in order that the chemical analogies should be pre-
served. Thus, the next known element after calcium (Ca = 40)
was titanium (Ti = 48). But if titanium were placed after calcium,
it would come in the third group, under aluminium, whereas its
properties indicate that the element is quadrivalent and ought
to go in the fourth group, under silicon :

Be 9 B 11 C 12 N 14

Mg 24 Al 27 Si 28 P 31

Ca 40 Ti 48 V 51

Zn 65 — As 75

470

INORGANIC CHEMISTRY

There was therefore a gap left in the third group, between calcium
and titanium. Two similar gaps were also left in the next period.
Mendeleeff predicted that these would be filled by unknown elements,
which he called ekaboron, eka- aluminium, and ekasilicon, respectively.

From the regularities of the atomic weights of the known elements
he was able to predict the atomic weights of the missing elements,
and from the positions in the table, he foreshadowed their pro-
perties in some detail. These predictions were brilliantly verified
by the discovery of scandium (Nilson, 1879), gallium (Lecoq de
Boisbaudran, 1875), and germanium (Winkler, 1886).

In the table below are given the predicted and observed pro-
perties of germanium ; these show how closely the predictions were
followed (Mendeleeff, " Principles of Chemistry," II, 27). It has been
said that these predictions could have been made without the
Periodic Law ; it may reasonably be asked why this had not in fact
been done.

EKASILICON (Es) ; predicted by
Mendeleeff, 1871.

Atomic weight 72.
Density 5-5.
Atomic volume 13.

Colour of element : dirty grey,
giving a white powder of EsO2
on calcination.

Metal will decompose steam
with difficulty.

Action of acids will be slight ;
that of alkalies more pro-
nounced.

Element will be obtained by
action of Na on EsO2, or
K2EsF6.

Oxide EsO2 will be refractory,
and have sp. gr. 4-7. Basic
properties of oxide less pro-
nounced than those of TiO2 or
SnO2, but more marked than
those of SiO2.

GERMANIUM (Ge), discovered by
Winkler, 1886.

Atomic weight 71-9.
Density 5-47.
Atomic volume 13-2.

Element is a greyish -white metal,
giving a white powder, GeO2, on
ignition.

Metal does not decompose
water.

Metal is not attacked by HC1 ;
it dissolves in aqua regia ;
aqueous KOH has no action,
but molten KOH oxidises Ge
with incandescence.

Element obtained by reduc-
tion of GeO2 by carbon, or of
K2GeF6 by Na.

Oxide GeO2 refractory ; sp. gr.
4-703 ; very feebly basic, al-
though indications of oxy- salts
are found.

xxiv CLASSIFICATION OF ELEMENTS, PERIODIC LAW

471

GERMANIUM (Ge), discovered by
Winkler, 1886.

Acids do not pp. hydroxide
from dilute alkaline solutions ;
from concentrated solutions,
acids or CO2 pp. GeO2 or meta-
hydroxide.

GeCl4 is a liquid, b.-pt. 96°,
sp. gr. 1-887 at 18°.

GeF4,3H2O is a white crystal-
line solid.

Ge(C2H5)4, b.-pt. 160°, sp. gr.
slightly less than that of water.

EKASELICON (Es); predicted by
Mendeleeff, 1871.

Hydroxide soluble in acids, but
solutions will readily hydro -
lyse with deposition of
meta-hydroxide.

Chloride EsCl4 will be a liquid,

b.-pt. below 100°, sp. gr. 1-9

at 0°.
Fluoride, EsF4, will not be

gaseous.
Organo -metallic compounds will

be formed ; e.g., Es(C2H5)4,

b.-pt. 160°, sp. gr. 0-96.

The reader should have no difficulty in following the predictions
of Mendeleeff from a consideration of the properties of the elements
silicon, tin, zinc, and arsenic, which are neighbouring elements in
the periodic table. The properties of gallium may also be inferred.

A complete new group, the zero group, was added to the table by
Ramsay, and numerous gaps in the lowest part of the table have
been filled in by the discovery of the radioactive elements (see
Chapter LI). The Periodic Law therefore points out the possi-
bility of discovering new elements ; it gives indications as to their
properties, and with what known elements they are likely to occur.
On the other hand, it shows that the number of possible new elements
is limited ; in particular, there are no new elements to be discovered
between helium (He = 4) and bromine (Br = 80), except possibly
in the transitional group or the zero group, because there is no
place for them in the table.

An important result of the periodic classification is the additional
confirmation it affords of the present values of the atomic weights,
and of the belief in the elementary character of the simple sub-
stances.

Difficulties in the periodic system. — The periodic classification,
in the form given to it by Mendeleeff, is not free from difficulties
and apparent contradictions. One of the most serious is the
inverted positions of three pairs of elements (A, K ; Co, Ni ; Te, I).
Again, it is very difficult to fit in the elements of the rare earths
(p. 461). The transitional elements occupy an exceptional position ;
attempts to include them in the other groups have not been
successful.

The arrangement into groups overlooks some chemical analogies,
such as those between boron and carbon, copper and mercury ;

472 INORGANIC CHEMISTRY CH. xxiv.

and also brings together elements which have little real analogy,
such as manganese and chlorine. The analogies between suc-
cessive elements in a period, pointed out by Mendeleeff , has, however,
often been neglected. Thus, the metals of the horizontal period :
V, Cr, Mn, Fe, Co, Ni, are chemically related ; and the sulphates,
RSO4, 7H20, of Mil, Fe, Co, Ni, Cu, Zn, are isomorphous.

The most remarkable difficulty, however, is the position of
hydrogen in the system. It is usually omitted altogether, but may
be given a whole period above the first complete period beginning
with helium. If it is placed in this period in Group I with the
alkali-metals, to which it shows resemblance in its electropositive
character and in forming an alloy with palladium, there must be a
number of other unknown elements in the period with atomic
weights between 1 and 4. The only other group in which a univalent
element could be placed is the halogen group, Group VII. But,
although hydrogen is a non-metal, can be replaced atom for atom
by halogens in organic compounds, and is a gas more difficult to
liquefy than fluorine, yet the period should then contain unknown
elements with atomic weights less than 1 (p. 468). However
placed, hydrogen occupies an exceptional position : its best situation
is probably at the head of Group I, on account of its electropositive
character, although some recent physical experiments place it
with the halogens.

EXERCISES ON CHAPTER XXIV

1. What steps would you take to ascertain (a) the atomic weight,
(6) the position in the periodic system, of a newly discovered metal?

2. Discuss the positions of (a) hydrogen, (b) potassium, (c) man-
ganese, (d) cobalt in the periodic system.

3. On what grounds was Mendeleeff able to predict the existence
and properties of gallium, scandium, and germanium ?

4. Discuss the reasons, other than the values of the atomic weights,
which justify the following pairs of elements being placed in the same
groups : beryllium and zinc ; iron and platinum ; sodium and copper ;
sulphur and chromium. With what other elements of other groups do
you consider each of these elements to be chemically related ?

5. Give a brief account of Prout's hypothesis. What bearing has it
on the interpretation of atomic weights ?

6. Discuss the position of iodine and tellurium in the periodic system.
What similar cases are known ?

7. What position is assigned to the rare-earth elements in the periodic
table ? What other alternative methods of classification would be
possible ?

CHAPTER XXV

SULPHUR AND ITS COMPOUNDS WITH HYDROGEN AND HALOGENS

Sulphur. — From its occurrence in the free state in Sicily, in the
centre of Roman civilisation, sulphur, or brimstone (German Brenn-
stein, i.e., combustible stone), has been known from antiquity.
The use of sulphur in medicine, and of the fumes of burning sulphur
in fumigation, are mentioned by Homer (c. 900 B.C.) ; the bleaching
of textile fabrics by the fumes was carried out at an early date.
The alchemists regarded sulphur as the principle of combustibility
and a constituent of metals (p. 29). The phlogistonists considered
it to be a compound of phlogiston and sulphuric acid ; the former
being evolved on burning, and appearing as a flame, whilst the acid
was left. Lavoisier (1777) pointed out that it should be regarded as
an element, and although Davy (1809) found that ordinary sulphur
always contains a little hydrogen, this was recognised as an impurity.

Sulphur occurs in Nature either free or in combination. Free
sulphur occurs in large quantities in Italy, in the volcanic regions
of Sicily, and in America, in the southern State of Louisiana. Less
important worked deposits occur in New Zealand in Whale Island,
in Texas, Chile, Russia, Iceland, and especially in Japan.

In 1884, 447,000 tons of sulphur were exported from Sicily, and
41,000 tons from the rest of the world. In 1913, Louisiana and Texas
alone produced 250,000 tons* Sicily 407,307 tons, Japan 58,452 tons,
and new Zealand (in 1914) 12,000 tons. More than 800,000 tons are
now said to be produced per annum in Louisiana.

Sicilian sulphur occurs stratified with marl, clay, and rock,
mostly gypsum, CaS04,2H2O, limestone, and quartz. It is found
occasionally in large, yellow, transparent crystals (Fig. 257), but
usually in crystalline masses, which are yellow or grey in colour.
Since fused sulphur deposits monoclinic crystals, which crumble
on standing to very small rhombic crystals, the origin of the
deposits, which contain large crystals, can hardly be igneous,
although rhombic crystals may have been deposited on very slow
cooling. The sulphur in the craters of extinct volcanoes is formed
by the interaction of volcanic gases, containing hydrogen sulphide

473

474

INORGANIC CHEMISTRY

CHAP.

FIG. 257. — Native Sulphur (British Museum).

and sulphur dioxide : 2H2S + S02 = 2H2O + 3S, probably derived
from pyrites. Since gypsum and calcium carbonate are always

found with the beds of sulphur,
the latter are assumed to be
the result of the reduction of
gypsum by organic matter and
bacteria : 2CaSO4 + 3C =
2CaCO3 -f 2S + CO2.

EXPT. 173. — Invert a jar of
sulphur dioxide over one of sul-
phuretted hydrogen. No action
occurs. Add a little water and
shake. The water becomes
turbid, from separation of sul-
phur, but no action occurs in the
gases. The latter must therefore
react in solution : 2H2S + SO2
= 2H2O + 3S.

Combined sulphur occurs in the form of metallic sulphides, many
of which are important ores of metals (i.e., serving for their extrac-
tion). E.g., lead sulphide, or galena, PbS ; zinc sulphide, or blende,
ZnS ; copper pyrites, Cu2S,Fe2S3 ; and iron pyrites, FeS2 (used as a
source of sulphuric acid). Hydrogen sulphide, H2S, occurs in
volcanic gases, and in some mineral springs, often with gaseous
carbon oxysulphide, COS. Sulphur dioxide, S02, also occurs in
volcanic gases. Some springs and rivers (Rio Canea and Rio
Vinagre, in America) contain free sulphuric acid, H2SO4. Large
masses of gypsum, or calcium sulphate, CaSO4,2H2O, and other
metallic sulphates, are common. Sulphur is a constituent of some
kinds of organic matter ; thus the blackening of silver spoons by'
eggs is due to the sulphur contained in the albumin of the latter.
It is found in certain bacteria, e.g., Beggiatoa alba, which are capable
of decomposing sulphur compounds in their life-processes. The
pungent principles of onions, garlic, horse-radish, and mustard are
organic sulphur compounds. Combined sulphur is present in hair
and wool, and in most animal and vegetable matter.

EXPT. 174. — Fuse a little hair with caustic soda in a test-tube.
Dissolve the cool mass in water, and pour on a silver coin. The latter
is at once turned black, through formation of silver sulphide, Ag2S.

The manufacture of sulphur. — Native sulphur, as it is dug in
Sicily, contains 15-25 per cent, of sulphur. It is stacked in lumps
in brick kilns, called calcaroni, built on sloping hillsides, with air-
spaces, and covered with powdered ore (Fig. 258). The ore is
kindled at the bottom, and the heat of combustion of about 30

xxv SULPHUR COMPOUNDS— HYDROGEN AND HALOGENS 475

per cent, of the sulphur serves to melt the rest, which flows off into
wooden moulds. The blocks so formed still contain 3-5 per cent,
of the original rock, and are exported to Marseilles for purification,
since fuel is too dear in Italy.

Improved methods of extraction are being introduced, e.g., the use
of the Gill kiln (1880), in which the heating is performed in closed brick
chambers, with six compartments in a circle, in the interior of which
a coke fire is kept burning. About 75 per cent, of the sulphur
is recovered. Payen and Gill (1867) also proposed to melt out the
sulphur with superheated steam ; the apparatus devised for this pur-
pose by Thomas (1869) is in use to a limited extent in the Romagna.

FIG. 258.— Calcaroni, or Sulphur Kiln

Sicilian sulphur is mostly refined at Marseilles, with the apparatus
shown in Fig. 259. The sulphur is fused in the iron pot, M, whence
it flows into the iron retort, G, heated over a fire. The sulphur
boils, and the vapour is conducted into a large brickwork chamber,
A. At first the vapour condenses on the cold walls as a light
yellow crystalline powder, called flowers of sulphur. As the walls
become hot, this melts (unless it is removed for sale), and runs down
as a liquid to the bottom, whence it is tapped off through 0 into
cylindrical moulds, to form roll sulphur, or brimstone. Such an
apparatus produces about two tons of refined sulphur per twenty-
four hours. Refining is also carried on at Romagna and Catania.

The Louisiana process of extraction is different. The deposit,
estimated at 40,000,000 tons, occurs below 900 ft. of clay, quicksand,
and rock. A boring is made to the deposit, and four concentric
pipes are sunk. Down the two outer pipes superheated water

476

INORGANIC CHEMISTRY

(155°) is pumped. This fuses the sulphur. Air is then forced down
the inner pipe, when an emulsion of water, molten sulphur, and air-
bubbles rises to the surface through the remaining annular space.
This passes to large wooden vats, where the sulphur, of 99-5 per
cent, purity, solidifies, and is ready for immediate use.

Sulphur was formerly prepared by distilling iron pyrites in clay
retorts : 3FeS2 = Fe3S4 + 2S (cf. 3MnO2 = Mn3O4 + O2) ; or by
roasting pyrites in kilns with a limited supply of air : 3FeS2 + 5O2 =
Fe3O4 + 3SO2 + 3S. It is more economical to burn the pyrites to
sulphur dioxide : 4FeS2 + 11O2 = 2Fe2O3 + 8SO2, and use this as a

source of sulph-
uric acid (p. 503).
Sulphur is formed
by heating metal-
lic sulphides to
1000° in carbon
dioxide: FeS +
COn = FeO +
CO+ S.

Sulphur from
alkali - waste. -

Some sulphur
is extracted from
Leblanc alkali-
waste (containing
insoluble calcium
sulphide, CaS), by
the Chance- Claus
process. A sus-
pension of the
waste in water
is treated with
limekiln gas,
containing car-
bon dioxide, in

large iron vessels called carbonators. Sulphuretted hydrogen
is evolved: (1) CaS + CO2 + H20 - CaC03 + H2S. The gas,
however, is too largely diluted with nitrogen (present in the
kiln gas) to pay for treatment. It is therefore passed into
a second carbonator where the sulphuretted hydrogen is
absorbed, the insoluble CaS passing into solution as calcium
hydrosulphide, Ca(HS)2 : (2) CaS + H2S = Ca(HS)2. When aU
the CaS in the first vessel is decomposed, this is cleaned out
and filled with fresh waste, and the connections are changed so
that the kiln gas passes directly into the second vessel. The

FIG. 259.— Refining of Sulphur by Distillation.

xxv SULPHUR COMPOUNDS— HYDROGEN AND HALOGENS 477

Ca(HS)2 is then decomposed : (3) Ca(HS)2 + CO2 + H2O =
CaCOg 4- 2H2S. The gas leaving the carbonator now contains,
for a given volume of nitrogen in the kiln gas passing through,
twice as much H2S as that from CaS in the first vessel, since equal
volumes of C02 are taken up in reactions (1) and (3). It is collected
in a large gas-holder over water covered with a layer of oil. This
gas is then mixed with air, and passed over porous oxide of iron
on a grating in the Claus kiln — a brickwork chamber, with large
brick condensing chambers and flues beyond. The oxide is
heated to start the reaction, which then proceeds automatically :
(4) 2H2S 4- 02 = 2H2O 4- 2S. The oxide of iron is unchanged, and
acts as a catalyst. Probably part of the H2S burns to SO.,, which
decomposes the rest : 2H2S + S02 - 2H2O 4- 3S— both S02 and
H2S are found in the waste gases after the sulphur has condensed
in the chambers : 35,000 tons of sulphur are recovered in England
annually by this process.

Sulphur from spent oxide.— Sulphur may also be extracted
from the spent oxide of the gasworks. Coal contains pyrites,
FeS2, the sulphur of which, during distillation in the manufac-
ture of gas, comes off chiefly as sulphuretted hydrogen, H2S,
and carbon disulphide, CS2 The former is removed by passing
the crude gas over hydrated oxide of iron, Fe(OH)3, mixed with
sawdust, in purifiers : 2Fe(OH)3 -f 3H2S == Fe2S3 + 6H20. When
the mass is no longer active, it is " revived " by exposure to air :
2Fe2S3 -f 302 4- 6H20 = 4Fe(OH)3 4- 6S. After these operations
have been repeated several times, the " spent oxide " contains
about 50 per cent, of free sulphur. It is then usually burnt in a
current of air to produce sulphur dioxide (p. 503). The sulphur
may be extracted from the material by solution in carbon disulphide,
but some tarry matter also dissolves.

Uses of sulphur. — Crude sulphur is used for making sulphur
dioxide (and thence sulphuric acid), and in the manufacture of
carbon disulphide (p. 710). Refined sulphur is used in medicine, in
the form of powder as a fungicide, and in the preparation of
gunpowder, matches, fireworks, and dyes. Sulphur is also used in
large quantities for vulcanising rubber.

Rubber, or caoutchouc, is a natural hydrocarbon obtained from the
juices of several tropical trees. In the untreated state, the elasticity
slowly disappears on warming and cooling, and to prevent this the rubber
is vulcanised (Hancock and Brockedon, 1847) by heating with 10-12
parts of sulphur at 140°, or by treating with a solution of sulphur in
sulphur chloride (p. 488). With larger quantities of sulphur (25-40
per cent. ) a hard mass of vulcanite, or ebonite, is formed.

For use in dressing vines (to prevent the growth of the fungus oidium),
sulphur is finely ground between millstones, and sieved through silk

478

INORGANIC CHEMISTRY

CHAP.

FIG. 260. — Crystals of Rhombic
Sulphur.

(170 meshes to the inch). By blowing a current of air through the
mill, the very finest particles ('• winnowed sulphur ") are carried off,
and are retained by cloth filters.

The allotropic forms of sulphur. — Sulphur exists in two common

crystalline forms : (1) rhombic, or a-sulphur
(Fig. 260), and (2) monoclinic, or
'/3-sulphur (Fig. 261). It also exists in
different amorphous forms, e.g., plastic
sulphur, or y-sulphur (now called
^-sulphur), and colloidal sulphur.

Native sulphur occurs in about
thirty-six crystalline varieties, all
belonging to the rhombic system
(p. 438). Rhombic or a-sulphur is pre-
pared by allowing a solution of sulphur

in carbon disulphide slowly to evaporate (p. 10), when pale-yellow,
transparent crystals are formed, giving a lemon-yellow powder.
The density of a-sulphur is 2-06, and its melting point is 112-8°.
It is insoluble in water, very slightly soluble in alcohol and ether,
freely soluble in carbon disulphide, sulphur chloride (S2C12), and
hot benzene and turpentine. Rhombic sulphur is the stable form at
the ordinary temperature, and all the other forms pass into a-sulphur
on standing. Roll sulphur consists almost entirely of rhombic
sulphur ; flowers of sulphur are principally composed of it (70 per
cent.), but contain also a white amorphous variety insoluble in
carbon disulphide.

Monoclinic or ^-sulphur was discovered in
1823 by Mitscherlich. Sulphur is dimorphous,
i.e., it exists in two distinct crystalline forms.
/3-sulphur is produced when fused sulphur is
allowed to crystallise.

EXPT. 175. — Half fill a beaker with small pieces
of roll sulphur, and heat gently on a sand-bath
till the whole is just fused. Allow to cool until a
crust forms on the surface. Make two holes in
this crust (one to admit air) with a pointed glass
rod, and pour the still liquid portion into a dry
porcelain dish. The inside of the beaker will be
found to be lined with beautiful, interlacing, transparent, flexible,
needle-shaped crystals (Fig. 261) of /3-sulphur, having a deeper yellow
colour than a-sulphur. On standing for a few days, the crystals
become opaque and brittle, and the colour becomes lemon-yellow.
The crystals now consist of aggregates of minute crystals of
a-sulphur, although the original monoclinic form is preserved by the
whole crystal, which is therefore called a pseudomorph. The gradual

)

FIG. 261..— Crystal of

Monoclinic Sulphur.

xxv SULPHUR COMPOUNDS— HYDROGEN AND HALOGENS 479

transition from one form to the other is readily followed by the
colour.

/3-Sulphur, when quickly heated, melts at 119-25°, and has a
density of 1-96. It is insoluble in water, but soluble in carbon
disulphide ; the solution on evaporation deposits a-sulphur.

Two other varieties of /3-sulphur, with slightly different angles, are
produced (i) by cooling in a freezing mixture a solution prepared by
heating sulphur with benzene, toluene, alcohol, or carbon disulphide in
a sealed tube ; (ii) by allowing an alcoholic solution of sodium sulphide,
saturated with sulphur, to stand in the air, when oxidation occurs,
and crystals grow on the surface. The crystals formed by process (i)
are thin flakes, called nacreous sulphur (French, nacre, mother-of-
pearl). Those -formed in
process (ii) contain hexa-
gonal plates (tabular
sulphur). Rhombohedral
and triclinic forms have P
also been described.

The transition point of
a- and ^-sulphur.—

Crystals of ^-sulphur, as
seen in EXPT. 175, slowly
change at the ordinary
temperature into minute
crystals of a-sulphur, and
become opaque. Crystals
of a-sulphur, on the other
hand, slowly become
opaque if heated above
96°, especially at 110°,
and pass into aggregates
of minute crystals of
/2-sulphur. The transformation of S^ into Sa is reversible ; below
96°, Sa is the stable form ; above 96°, 8/3 is stable. This tem-
perature, 96°, is called the transition temperature or transition point
of sulphur. At the transition temperature, both crystalline forms
are in equilibrium, Sa ;=± S^ .

Substances, such as sulphur, which exist in two forms, one of which is
stable below a certain temperature and the other stable above it, are
called enantiotropic ; substances like iodine chloride, which exist only
in one stable form, the other forms being unstable in all circum-
stances, are called monotropic (Greek mono, one ; enantios, opposite ;
tropos, form).

Equilibrium between a- and ^-sulphur.— a-Sulphur, /3-sulphur,

Vapour

FIG. 262. — Phase Rule Diagram for Sulphur.

480 INORGANIC CHEMISTRY CHAP.

liquid sulphur, and sulphur vapour are different phases of sulphur
(p. 7), and according to the Phase Rule (p. 106) they ought to
coexist under certain conditions of temperature and pressure.
In Fig. 262, OP is the vapour-pressure curve of Sa, i.e., it represents
the pressures of sulphur vapour in equilibrium with solid Sa at
various temperatures. QZ is the vapour-pressure curve of liquid
sulphur. The point R, which is the point of intersection of OP
and QZ, defines a temperature and pressure at which Stt, liquid S,
and S-vapour coexist in equilibrium. It is the melting point of
Sa under its own vapour pressure, about 113°, and is a triple point
(3 phases: Sa, liquid, vapour, in equilibrium, cf. p. 92). PQ is
the vapour-pressure curve of /2-sulphur, meeting QZ at Q. Q
therefore defines the temperature and pressure at which 80, liquid S,
and S-vapour are in equilibrium — it is another triple point, viz., the
melting point of /8-sulphur under its own vapour pressure, 120°.

P Q also crosses OP at P. P is the triple point at which Sa, 8/3,
and S-vapour coexist ; it is the transition point of a- and /3-sulphur,
96°. Below 96°, a is stable and/3 unstable ; above 96°, ft is stable
and a unstable. But S/3 may exist in a metastable condition below
96°, because the change S^ -> Sa takes place only slowly. The
prolongation of QP to Y expresses this fact, PY being the vapour-
pressure curve of Sft at temperatures below 96°. The melting
points of a- and /3-sulphur are raised by pressure (cf. ice, p. 91),
but at different rates. This is represented by two lines, starting
from P and Q with different slopes (3T/Sp), and meeting ulti-
mately at ${151° ; 1228 atm.), where Sa, Sft, and liquid S are in
equilibrium. Above this point S/s cannot exist, and the region of
existence of Sp is confined to the area PSQ. The areas defining
the regions of existence of Stt, liquid, and vapour are marked. The
point R is inside the region of S ftt hence the melting point of Sa is
a metastable point ; it can be realised only because the change
Sa -> Sp is so slow that fusion of the former, at its appropriate
melting point (112-8°), takes place before the change Sa -> Sp,
which begins at 96°, has proceeded appreciably. If Sa is kept
at a temperature below 112-8° but above 96° for a long time, and
then heated, it will not melt at 112-8°, but at 120°, since it is now
all converted into Sft .

Amorphous sulphur. — EXPT. 176. — If pieces of roll sulphur are heated
in a flask they melt to a clear yellow liquid at 112-8°. This is called
S\. On cooling rapidly in water, S« is produced, completely soluble
in CS2. If the temperature is now gradually raised, the liquid, at first
quite mobile, suddenly becomes very viscous, and its colour darker
yellow at 180-190°. At 230° the liquid is black and viscous. This
form of the liquid is called SM. Beyond 230° the viscosity decreases,
but the colour remains dark, and the sulphur finally boils at 444°.

xxv SULPHUR COMPOUNDS— HYDROGEN AND HALOGENS 481

If the boiling liquid is poured into cold water it forms soft, sticky,
rubber-like, transparent threads, called plastic sulphur, y-sulphur, or
/u-sulphur.

Sfj. has a density of 1 -96 ; it is insoluble in carbon disulphide.
On standing for a few days the threads form an opaque, brittle
solid, lemon-yellow in colour, consisting mainly of Sa- About 34
per cent, of the solid is still insoluble in carbon disulphide, and
consists of S/j.. At 100°, the change from viscous liquid to solid
takes place in an hour.

It has been found that the darkening in colour is due to organic
impurities, and that S/x (plastic sulphur) is only formed if slightly
impure sulphur, which has been exposed to air and contains sulphuric
acid, is used. If ammonia gas is passed through the melted sulphur,
no plastic sulphur is formed on further heating. In the liquid, S\ and
Sju, exist in equilibrium at various temperatures : SA ^± S^. E.g., the
percentages of S^ are : at 120°, 3-6 ; 160°, 11 ; 444-7°, 30. A trace of
iodine stabilises S^-

Two other varieties of sulphur have recently been described. Sn is
obtained when sulphur is heated above the melting point and rapidly
cooled ; its solution in carbon disulphide has a deep yellow colour.
84, is produced when concentrated hydrochloric acid at 0° is added to a
cold solution of sodium thiosulphate and the mixture shaken with
toluene. After a short time orange -yellow crystals of S<£ separate
from the toluene, having a distinct form and solubility. The solutions
of 84, are yellow, but not so strongly as those of STT-

If 2 parts of flowers of sulphur are boiled with 13 parts of water
and 1 part of lime slaked with 3 parts of water, the clear liquid
decanted is deep reddish-yellow in colour, and contains poly-
sulphides of calcium, CaS-Sn, e.g., CaS5. The early Greek alchemist
Zosimus refers to this liquid as " the divine water " (thion hudor),
or " the bile of the serpent." Pliny stated that it was used by the
Romans to give a dark gloss to silver — " oxidised " silver is so made
to-day. The Latin Geber stated that if an acid be added to the
liquid, a very offensive smell (sulphuretted hydrogen) is noticed,
and a fine white precipitate of sulphur is formed. This is called
milk of sulphur (lac sulphuris), and is prepared for pharmaceutical
purposes. It is soluble in carbon disulphide.

If the thion hudor is precipitated with dilute sulphuric a^eid, the
resulting sulphur may contain calcium sulphate ; this is left as a residue
on burning a little of the sulphur on platinum foil.

Another amorphous variety of sulphur remains as a pale yellow
powder when " flowers of sulphur " are treated with carbon disul-
phide. This form also separates when a solution of sulphur in

I I

482 INORGANIC CHEMISTRY CHAP.

carbon disulphide is exposed to sunlight, or on the decomposition of
sulphur chloride by water.

Colloidal sulphur is formed in the preparation of milk of sulphur :
the filtered liquid is a turbid emulsion of minute drops of liquid
sulphur soluble in CS2, and doubly-refracting (liquid crystals). If a
solution of sodium thiosulphate (" hypo ") is acidified, it quickly
forms a turbid colloidal suspension of minute solid crystals of
sulphur, insoluble in CS2. If the milky liquid obtained by pass-
ing sulphuretted hydrogen into a solution of sulphur dioxide is
evaporated, a gum-like residue is left, part of which is soluble
and part insoluble in CS2. These varieties of colloidal sulphur
were called S-sulphur by Debus (1888).

Sulphur vapour. — Sulphur boils at 444-7°, and forms a deep red
vapour, which, when strongly heated, becomes yellow. Dumas
(1832) found the vapour density at 524° to be 96, which corresponds
with the molecule S6 ; at higher temperatures the density fell, and
Dumas thought the molecules S4 and S2 were formed. Biltz (1901),
working with a wider range of temperatures, found the following
densities: 468°: 113 (higher than S7) ; 524°: 102 (higher than
Dumas' figure) ; red heat : 32-2 (S2). He concluded that at
lower temperatures the molecule is S8, but this is partially disso-
ciated even at the boiling point. There was no evidence of a
constant density over any range of temperature, and Biltz con-
sidered that the molecule S8 broke up at once into S2 : S8 ^ 4S2.
The lowering of vapour pressures of CS2 and S2C12 containing
dissolved sulphur gives the formula S8 for the latter. Bleier and
Kohn (1900) found that the vapour density rises when the boiling
point is lowered by diminished pressure. At 214° (2 mm. pressure)
the density corresponds with 7-J atoms in the molecule. Preuner
(1903) considers that S6 and S4 also occur in the vapour : Sg^
S6 -f S2 = S4 -f 2S2 ^ 4S2. Nernst found that 45 per cent, of
the S2 molecules were broken up into atoms at 1900-2000°:
S2 — 2S.

Pure sulphur. — H. B. Baker purified sulphur by heating the vapour
with S2C12 at 450°, when the hydrogen present as impurity forms
H2S, which reacts with S2C12 to form HC1 and S. The S2C12 and HC1
were removed by heating in vacuo, and the sulphur left was so pure
that it could be distilled unchanged in oxygen dried over phosphorus
pentoxide (cf. p. 704).

Compounds of sulphur with hydrogen. — Sulphur forms a gaseous
compound H2S, hydrogen sulphide, or sulphuretted hydrogen, analogous
to water, H2O. In a series of analogous compounds of related
elements, the boiling point rises with the atomic weight of the
element ; hence Vernon infers that water, which should boil at a
lower temperature than H2S, must be associated, (H20)n (cf.

xxv SULPHUR COMPOUNDS— HYDROGEN AND HALOGENS 483

p. 201). Two liquid hydrogen persulphides, H2S2, H2S3, are known
(cf. H202).

Sulphuretted hydrogen, H2S. — When hydrogen is passed over
boiling sulphur in a bulb-tube, the issuing gas contains a small
amount of sulphuretted hydrogen (1-2 per cent.), and blackens
lead acetate paper, owing to the formation of lead sulphide, PbS
(ExPT. 177). If pure sulphuretted Iwdrogen is heated, partial
decomposition occurs, with deposition of sulphur. The reaction
is therefore reversible : H2 -f- S — H2S, and a state of equilibrium
is reached. At 310°, the combination H2 -f- S = H2S is almost
complete after a week, whilst if a stream of powerful sparks is
passed through H2S, sulphur is deposited, and the gas left, which
occupies the same volume, is nearly
pure hydrogen.

Traces of sulphuretted hydrogen are
formed when sulphur is boiled with
water : 2H2O + 2S ±^ 2H2S -f O2.
The gas is also formed when heavy
naphtha (sp. gr. 0-9) is dropped into
boiling sulphur in a flask ; part of the
hydrogen in the hydrocarbon is
substituted by sulphur (S replaces 2H),
and forms H2S (cf. p. 398).

Sulphuretted hydrogen is usually
prepared by the action of dilute
sulphuric, or hydrochloric, acid on
ferrous sulphide : FeS + H2SO4 =
FeS04 + H2S. The reaction is
carried out in a Kipp's apparatus
(Fig. 263), so that the supply of the
gas, which has a most unpleasant odour and is a blood-poison,
may be interrupted at will. On account of the invariable presence
of free iron in the ferrous sulphide, the gas so prepared contains
hydrogen, which does not interfere with its use in qualitative
analysis.

Hydrogen sulphide free from hydrogen is obtained by boiling
powdered native antimony sulphide (stibnite] with concentrated
hydrochloric acid : Sb2S3 + 6HC1 = 2Sb013 -f 3H2S. The pure gas
is also formed by treating calcium or magnesium sulphides with
acid : CaS + 2HC1 = CaCl2 -f H2S ; or by heating to 60° a solution
of magnesium hydrosulphide, obtained" by passing the impure
gas from FeS through magnesia suspended in water :

MgO + 2H2S ~ Mg(HS)2 + H20.

This is a reaction of hydrolysis (p. 360).

i i 2

FIG. 263.— Kipp's Apparatus for H2S.

484 INORGANIC CHEMISTRY CHAP.

When concentrated sulphuric acid is heated with zinc, sulphuretted
hydrogen is formed : 4Zn -j- 5H2S04 = 4ZnS04 + 4H2O -f- H2S.

The action of acids on sulphides. — Ferrous sulphide is very slightly
soluble in water, and is almost wholly ionised at the great dilution*:
(1) FeS^Fe" + S". The solubilities of some sparingly soluble
sulphides are given below, in gm. mols. per litre :

MnS2-6x lO-8 PbS 2 x ID-*4 Ag2S 2 x H)-"

FeS 6 x 10-10 CdS 7 x 10-^ CuS N)-»

ZnS 7 X 10-12 Bi2S3 3 X lO'19 HgS 3 x 10-2*(?)

Sulphuretted hydrogen is a weak dibasic acid : (2) H2S ^H' -f HS'
^ 2H' -1- S", and the second stage of its ionisation, to S", is very
slight. The concentration of S" formed in (2) is therefore still less
than that formed in consequence of (1), and on adding a strong acid
the H* ions of the latter combine with the S* ions of the sulphide
to form H2S until the concentration [S*] in the solution is reduced
to a value compatible with (2). The solubility product of H2S is
then exceeded, the gas is formed, and escapes from the liquid.
From the equation : [S"] X [H*]2 == const., we see that, since
[S"] from the trace of dissolved sulphide is very small, [H'] must be
large in order to produce the value of the product corresponding
with a saturated solution of H2S. If the sulphide is very sparingly
soluble (e.g., CuS, HgS), the necessary concentration of [H'] can-
not be produced, even by strong acids, and these sulphides do
riot dissolve in the latter. When treated with nitric acid (which
causes oxidation, with separation of sulphur, or forms sulphuric
acid) they dissolve. In the case of cadmium sulphide, CdS, the
H2S accumulating stops the reaction before solution is com-
plete, and very strong acid must be used, or the H2S must be
removed from the liquid by boiling, or by a current of air.

Properties of sulphuretted hydrogen. — The gas may be collected
over hot water ; it is appreciably soluble in cold water (4-37 vols.
at 0°, 3-58 vols. at 10°, 2-9 vols. at 20° ; 1 vol. of alcohol at 15°
dissolves 8 vols. of H2S). It may also be collected by displace-
ment, since its density is 1-2 (air = l). It attacks mercury
slowly.

Hydrogen sulphide is a colourless gas with a powerful odour of rotten
eggs (decaying albumin evolves H2S), and is poisonous ; it liquefies
at — 61 -8°, the vapour pressure at 12° being 15 atm. At lower
temperatures it forms a transparent solid, melting at — 83°. The
critical temperature of H2S is 100°, the critical pressure 90 atin.

The aqueous solution is a feeble acid ; the gas is com-
pletely expelled by boiling, and on standing in the air the
solution becomes turbid, owing to oxidation and deposition
of sulphur : 2H2S -f- O2 = 2H2O -f 2S. This is retarded by the
addition of glycerin (cf. p. 494). In decinormal solution 0-07 per

xxv SULPHUR COMPOUNDS— HYDROGEN AM) HALOGENS 485

cent, is ionised to H' -f HS' ; the further stage, to S", proceeds
only very slightly : H2S ^ H' + HS' -^ 2H' + S".

The gas is decomposed by sparks, or by a heated platinum spiral :
H2S ^ H2 + S. It is also decomposed by heated sodium, tin, or
lead, giving its own volume of hydrogen, and sulphides of the
metals : H2S -j- Sn = H2 -f- SnS. Its density is 17, /. mol. wt.
= 34. Of this, the hydrogen H2 accounts for 2, /. wt. of S — 32,
which is the atomic weight ; hence the formula of sulphuretted
hydrogen is H2S.

If a jar of chlorine is inverted over one of sulphuretted hydrogen,
and the plates are removed, the gas deposits sulphur : H2S + C12 =
2HC1 -j- S (EXPT. 178). If a sofution of sulphuretted hydrogen is
treated with a large excess of chlorine water, the solution contains
sulphuric acid : S + 4H20 + 3C12 = H2SO4 -f 6HC1. Hydrogen
sulphide, on account of the ease with which it is oxidised, is a
reducing agent, and is used for this purpose, in aqueous or alcoholic
solution.

The gas burns in air or oxygen with a blue flame, and owing to
the high temperature it is completely dissociated in the interior
of the flame ; the latter deposits sulphur on a cold porcelain dish.
If the gas in a glass cylinder is ignited at the mouth, a deposit of
sulphur is formed on the inside of the jar, owing to the deficiency
of oxygen : 2H2S + O2 = 2H20 -f 2S (ExPT. 179). With a plen-
tiful supplv of oxygen, sulphur dioxide is formed : 2H2S -f 3O2 =
2H20 + 2S02. A mixture of 2 vols. of H2S and 3 vols of 02 explodes
violently on ignition.

The gas, or its solution (e.g., mineral waters), may be detected by
the black coloration, due to lead sulphide, PbS, produced with lead
acetate. If alkali sulphides are present, they give a purple colour
with a freshly-prepared solution of sodium nitroprusside ; this is
not produced by free H2S. The gas decomposes sulphuric acid and
calcium chloride, and must be dried over phosphorus pentoxide :
H2S04 + H2S = S + SO2 + 2H20. It is absorbed by caustic
soda.

Precipitation of metallic sulphides. — Sulphuretted hydrogen pre-
cipitates sulphides of metals from many solutions of salts of the
latter. These sulphides often have characteristic colours, and H2S
is used as a reagent in qualitative analysis.

EXPT. 180. — Pass a current of sulphuretted hydrogen through a
series of wash-bottles (Fig. 264) containing solutions of lead acetate ;
copper sulphate ; mercuric chloride ; arsenious oxide in dilute hydro-
chloric acid ; antimony chloride ; cadmium sulphate (a) slightly acidified
(b) strongly acidified, with HC1 ; notice the effects produced.

Many sulphides are precipitated from solutions acidified with
hydrochloric acid : copper, lead, mercuric and bismuth salts, all give

486

INORGANIC CHEMISTRY

black sulphides ; cadmium and arsenic give yellow sulphides,
CdS, As2S3 ; antimony gives an orange-red sulphide, Sb2S3 ; tin
(stannous) a brown sulphide, SnS.

In some cases metals are precipitated only in alkaline solutions.
An alkali sulphide, e.g., ammonium sulphide, may be used.

EXPT. 181. — Add ammonium chloride and excess of ammonia to
solutions of zinc sulphate, manganous sulphate, and nickel sulphate in
bottles, and pass a stream of H2S through the liquids. Note the colours
of the precipitates (ZnS, MnS, NiS).

The precipitation of sulphides of metals may be considered from
the same point of view as their solution in acids, (i) If the sulphides
are very sparingly soluble (PbS, CuS, HgS, As2S3, Sb2S3. etc.) the

Fm. 264.— Precipitation of Sulphides of Metals.

concentration of S* ions formed from them is never large enough,
even with relatively high concentrations of H' ions, to give an
ionic product [H']2 x [S"j exceeding the solubility product of
H2S, so that the latter cannot be formed. In other words, the
sulphides are precipitated even in the presence of acids, (ii) Cad-
mium sulphide, CdS, occupies an intermediate position. If the
acid concentration is greater than 0-3 N it is not precipitated,
(iii) Sulphides of other metals (FeS, ZnS, MnS) are precipitated in
alkaline solution, because then no H" ions are formed :

ZnS04 + (NH4)2S = ZnS + (NH4)2S04.

(iv) The metals of the alkalies and alkaline earths are not preci-
pitated, because their sulphides are soluble in water (Na2S, K2S), or
in a solution of sulphuretted hydrogen (GaS -j- H2S ±^ Ca(SH)2).
(v) Aluminium and chromium salts give precipitates of hydroxides

xxv SULPHUR COMPOUNDS— HYDROGEN AND HALOGENS 487

with ammonium sulphide, since their sulphides are completely
hydrolysed by water :

2A1C13 + 3(NH4)2S -f 6H,0 - 2A1(OH)8 + 6NH4C1 + 3H2S ; or
2A1 ' ' + 3S""+ 6H20 = 2A1(OH)8 + 3H2S.

Hydrogen persulphides. — If an acid is added to the yellow solution
of polysulphides of calcium (p. 481) which contains CaS2 and
probably CaS5, sulphuretted hydrogen is evolved, and white col-
loidal sulphur is formed, slowly depositing as milk of sulphur :
CaS2 + 2HC1 =* CaCl2 + H2S -f S. Scheele (1777) found, how-
ever, that if the calcium sulphide solution is poured in a thin stream
into cold; fairly concentrated hydrochloric acid, with constant
stirring, a yellow oil separates, which Thenard (1832) regarded as
hydrogen persulphide, H2S2, analogous to H202. A piece of litmus
paper placed in the liquid is bleached (ExFT. 182). The oil, which
may be separated by a tap-funnel, has a pungent smell, its sp. gr.
is 1 -7 ; it is soluble in benzene and carbon disulphide, but is sparingly
soluble in, and is decomposed by, alcohol. It slowly decomposes
spontaneously, especially on warming, into sulphuretted hydrogen
and a residue of sulphur. If sealed up in a bent tube, liquid H2S col-
lects in one limb, cooled in a freezing mixture, and sulphur remains
in the other. The formula of the oil is, therefore, H2S#, but its
composition is variable, since the sulphur formed on decomposition
dissolves in the remaining persulphide. Some chemists considered
it to be H2S5, but more recent work shows that it is a solution of
sulphur in H2S2 and H2S3.

Sabatier (1885) separated the crude persulphide into fractions by
distillation under reduced pressure ; under 40-100 mm. pressure the
chief fraction has a composition intermediate between H2S2 and H2S3.
Sabatier concluded that it was H2S2 + dissolved sulphur. Bloch and
Holm (1908), by using glass vessels treated with hydrochloric acid
to remove alkali (which decomposes the persulphide), separated the
crude oil by distillation under reduced pressure into two volatile frac-
t'ons. In the first receiver, hydrogen trisulphide, a pale yellow liquid,
sp.gr. 1-496, b.-pt. 43-50°/4-5 mm., m.-pt. — 52-53°, collected;
and in a further, strongly cooled receiver, hydrogen disulphide, H2S2,
a yellow liquid, sp. gr. 1-376, b.-pt. 74-75°, quickly decomposed by
water and alkalies, was obtained. These are supposed to undergo
intramolecular change, so that the liquids contain different molecules in
equilibrium (p. 497) :

S:S< ' ±=: HS-SH ; S:S:S<1i ±=r S:SH>SH — HS'S'SH.

\±1 \±1

Halogen compounds of sulphur. — The following halogen com-
pounds of sulphur are known :

SF6; S2C12, SC12(?), SC14 ; S2Br2.

488 INORGANIC CHEMISTRY CHAP.

Sulphur burns spontaneously in fluorine, prodifcing a colourless
gas, sulphur hexafluoride, SF6 (Moissan and Lebeau, 1900). This
is of interest as an example of the maximum valency of sulphur,
viz., 6. The gas is chemically inert, like nitrogen, but is decom-
posed by boiling sodium : SF6 -f- 8Na = Na2S + 6NaF. Its
relative density is 73 ; it solidifies at — 55°. Even fused caustic
potash and ignited lead chromate or copper have no action upon
it ; H2S is decomposed by SF6, with formation of HF and S.

Sulphur monochloride, S2C12, is prepared by passing dry chlorine
over sulphur fused in a retort (Thomson, 1804). A reddish-yellow
liquid distils over into a cooled receiver (Fig. 265). By rectification
of this a clear amber-coloured liquid, sp. gr. 1-706, boiling at 138°,

FIG. 265. — Preparation of Sulphur Monochloride.

is obtained. This solidifies at — 80°. Sulphur monochloride has
a vapour density of 67-6, which corresponds with S2C12 (A =67-0).
Sulphur monochloride fumes in moist air, and has a most disagree-
able pungent odour. The stoppers of bottles in which it is kept
become coated with sulphur owing to this hydrolysis :

2S2C12 + 3H20 = 4HC1 + 2S + H2S203.

The liquid itself is only slowly decomposed by water ; hydrochloric
acid and sulphur are formed, together with various oxy-acids of
sulphur (e.g., thiosulphuric acid, H2S203, pentathionic acid,
H2S5O6, etc.). Metals decompose it on heating, forming chlorides
and sulphides. S2C12 dissolves sulphur readily (66 per cent.), and
the solution is used in vulcanising rubber (p. 477)

xxv SULPHUR COMPOUNDS— HYDROGEN AND HALOGENS 481)

If SaCla is saturated with C12 at — 22°, a yellowish-brown liquid is
formed, which is sulphur tetrachloride, SC14. This freezes to a yellowish-
white solid, melting at — 31°. On taking the liquid out of the freezing
mixture, it decomposes into S2C12 and C12. Stable double compounds,
e.g., 2A1C13,SC14, are known. The liquid formed by saturating S2C12
with C12 at the ordinary temperature was considered to be the dichloride,
SC12, but is probably a solution of SC14 in S2C12. Double compounds
corresponding with the dichloride, e.g., AsCl3,SCl2, however, are
known. Sulphur monobromide, SBr, or S2Br2, is a garnet-red liquid,
b.-pt. 57~/0'2 mm., m.-pt. -- 46°, obtained by heating sulphur with
bromine in a sealed tube.

EXERCISES ON CHAPTER XXV

1. How would you prove experimentally that all the different modi-
fications of sulphur consist of the same chemical element ? It is some-
times said that, like oxygen and ozone, they contain different amounts
of energy : how could this be tested ?

2. From what sources is sulphur obtained ? What varieties of
sulphur exist, how are they prepared, and what are their properties ?

3. How is pure hydrogen sulphide prepared ? Give a general account
of the action of the gas on solutions of metallic. salts. How is its for-
mula established ?

4. How are the persulphides of hydrogen obtained ? Point out the
resemblances and differences between hydrogen peroxide, H2O2, and
hydrogen persulphide, H2S2.

5. Describe the preparation and properties of the halogen compounds
of sulphur. What light do these compounds throw on the valency of
sulphur ?

6. Discuss the allotropy of sulphur from the point of view of the
phase rule.

CHAPTER XXVI

THE OXYGEN COMPOUNDS OF SULPHTR

Oxygen compounds of sulphur. — The following oxides 01 sulphur
are known :

Sulphur sesquioxide, S2O3 : possibly the anhydride of hyposul-

phurous acid, H2S2O4 ;

Sulphur dioxide, SO2 : the anhydride of sulphurous acid, H2SO3 ;
Sulphur trioxide, SO3 : the anhydride of sulphuric acid, H2SO4 ;
Sulphur heptoxide, S2O7 : the anhydride of persulphuric acid, H2S2OV

A large number of oxy-acids of sulphur are known, either in the
free state or in salts :

Hyposulphurous acid, H2S2O4 Dithionic acid, H2S2O6

Sulphurous acid, H2SO3 Trithionic acid, H2S3O6

Sulphuric acid, H2SO4 Tetrathionic acid, H2S4O6

Thiosulphuric acid, H2S2O3 Pentathionic acid, H2S5O6

Pyrosulphuric acid, H2S2O7 Hexathionic acid, H2S6O6

Persulphuric acid, H2S2Og
Permonosulphuric acid,

or Caro's acid, H2SOt

SULPHUR DIOXIDE.

Sulphur dioxide, SOj.— Homer (B.C. c. 1100-900) refers to the use of
burning sulphur in fumigation, and Pliny states that the fumes were
also used for purifying cloth (i.e., bleaching). The alchemists thought
the pungent fumes were oil of vitriol but Stahl (1702) showed
that they gave peculiar salts with alkalies, and since they stood
halfway between sulphuric (vitriolic) acid and sulphur (the la*
regarded as sulphuric acid -f- phlogiston), he called the acid phlogis-
ticated vitriolic acid. Priestley (1774) obtained the pure gas by
heating concentrated sulphuric acid with mercury, and collected
it over mercury. He called it vitriolic acid air. Its composition
was ascertained by Lavoisier in 1777 by burning sulphur in a
measured volume of oxygen ; it is sulphur dioxide, SO2.

The combustion of sulphur. — When roll sulphur is heated in air
it fuses, and as the temperature rises a very genth- combustion

c... xxvi THE OXVCHX C'OMPorNDS OF Sl'Ll'llUi

491

begins, accompanied by a faint glow, visible only in a dark room.
This is due to the oxidation of sulphur vapour, which comes
off appreciably at about 230°. At about 360° in air (275-280° in
oxygen) the sulphur ignites, and burns with a blue flame, producing
sulphur dioxide, S02, and a little solid sulphur trioxide, SO3, which
renders the gas cloudy. Sulphur dioxide also becomes cloudy in
a strong beam of light (Tyndall effect, p. 7), owing to the decom-
position into fine particles of S03 and sulphur : 3SO2 = 2SO3 -f S.
The reaction is reversible, and the gas becomes clear again on stand-
ing in the dark.

Sulphur burns in a con-
fined volume of oxygen or
air without causing appreci-
able change of volume, i.e., II flj
Mtlphur (//o. nWr contains ite
oioi rol-uwc c>f o.nigcn (Priest-
Icy. 1772).

EXPT. 183. — A small piece
of sulphur lying in a platinum
spoon is ignited in ilry oxygen
gas confined over (In/ mercury
in the apparatus shown in
Fig. 2lH>. by means of apiece
of fine platinum wire heated
electrically in contact with
the sulphur. When the
apparatus is cool it is found
that t hi' mercury levels are
practically unchanged. There
is a very slight contraction due
(i) to the formation of a little
solid SO3 ; (ii) to the greater
compressibility of SOa as
compared with ().,.

Fu;. 200. — Volumetric Composition of Sulphur
Dioxide.

The normal density of sulphur dioxide is 2-9266, hence its relative
density (H = 1) is 2-9266 ^ 0-09 = 32-5. The molecular weight
is therefore approximately 32-5 x 2 = 65-0. But the above
experiment shows that the molecular weight of the gas contains a
molecular weight of oxygen, O2 = 32, hence its formula is S*O2.
The remainder, 65 — 32 = 33, is the weight of sulphur. But the
atomic weight of sulphur is 31-81, hence the formula is SO2.
The exact molecular weight is therefore 31*8 -f- (2 X 15*88) =

<>:*•:><; (ll - i).

Sulphur dioxide is prepared on UK* large scale by the combustion

492

INORGANIC CHEMISTRY

2H2O

of native sulphur, or of iron pyrites, in a current of air in special
burners (cf. p. 503). It is used in bleaching wool and straw and
as a disinfectant. The largest proportion is used directly in the
manufacture of sulphuric acid.

Preparation of sulphur dioxide. — In the laboratory the gas is
usually made by the reduction of sulphuric acid. If concentrated
sulphuric acid is heated with copper, mercury, silver, sulphur, or
charcoal, it is reduced, and sulphur dioxide is formed :

2H2S04 + Cu = 2H2O + CuS04 + SO2 ;
Hg + 2H2S04 = HgS04 -f 2H2O + SO2 ;
2Ag + 2H2S04 = Ag2S04 + 2H20 + SO2 ;
S 2H0SO4 = 3S02 + 2H20 ;

C02.

EXPT. 184. — About 100 gm.
of copper turnings are covered
with concentrated sulphuric
acid in a flask fitted with a
thistle funnel (Fig. 267), and
heated on wire gauze. The
mixture becomes very dark,
and gas is evolved with effer-
vescence. When this occurs
the flame is removed. The
gas is collected by downward
displacement (density 2-26
times that of air), or over
mercury. After cooling, the
residue in the flask is warmed
with water, the solution
filtered, evaporated, and set
aside. Deep blue crystals of

FIG. 267. — Preparation of Sulphur Dioxide.

copper sulphate, CuSO4,5H2O (blue vitriol), separate.
Some black, insoluble cuprous sulphide, Cu2S, is
always produced in the reaction (p. 812).

A more convenient method of preparation is to
drop concentrated sulphuric acid into a saturated
solution of sodium hydrogen sulphite (" bisul-
phite") : NaHS03+H2S04 = NaHS04 + H20 +
S02.

The gas is most conveniently obtained from
the liquid, which is sold in glass siphons (Fig.
268). By inverting these, the Liquid is delivered.

Properties of sulphur dioxide. — Sulphur dioxide
is a colourless gas, 2-264 times heavier than air.

FIG. 268.— Liquid

xxvi THE OXYGEN COMPOUNDS OF SULPHUR !<>;{

It has a choking smell, well known as that of burning sulphur,
and is poisonous. It does not support combustion in the ordinary
sense, but potassium takes fire spontaneously in the gas :

4K + 3S02 = K2SO3 (sulphite) + K2S2O3 (thiosulphate).

Finely-divided tin and iron also burn in the gas when warmed, form-
ing mixtures of oxides and sulphides. A little lead dioxide in a
deflagrating spoon, when warmed and introduced into the gas,
becomes incandescent, and forms white lead sulphate : Pb00 + SO0
= PbS04.

When exposed to 2 aim. pressure at 15°, S02 forms a colourless
liquid, b.-pt. — 10-09° ; on rapid evaporation, this freezes to a
snow-like solid, m.-pt. --76°. The critical temperature is 152-7°,
the critical pressure
77-95 atm. The
liquid, sp. gr. 1-434
at 0°, readily dis-
solves iodine, sul-
phur, phosphorus,
resins, and some
salts. The solutions
of the latter con-
duct the electric
current feebly, so
that the solvent
has slight ionising
properties.

The ionising power
of a solvent depends
on its dielectric con-
stant. Water, with

a dielectric constant of 81, is a good ionising solvent ; benzene (2-3)
and sulphur dioxide (13'75) are poor ionising solvents; alcohol (26)
occupies an intermediate position.

EXPT. 185. — Liquid S02 is easily prepared by passing the gas through
a glass spiral immersed in a mixture of pounded ice and salt (Fig. 269).
The liquid is collected in a strong tube with the neck drawn off, immersed
in freezing mixture. The neck may be sealed whilst the tube remains
cooled, and the liquid preserved.

Sulphurous acid. — Sulphur dioxide is freely soluble in water,
forming a liquid smelling strongly of the gas, and acid to litmus.
It probably contains the unstable sulphurous acid, H2S03, but the
latter has never been isolated. On warming, sulphur dioxide is
evolved. When the saturated solution is strongly cooled, crystals

FIG. 269.— Liquefaction of SO2 by Cooling.

494 INORGANIC CHEMISTRY CHAP.

of the hydrate, S02,7H20, separate. The solution when heated in
a sealed tube at 150° deposits sulphur :

3H2S03 = 2H2S04 + H20 + S (cf. 3S02 = 2S03 + S).

The solution of sulphurous acid possesses bleaching properties ;
moistened wool, straw for hats, and other materials which would be
injured by chlorine, are bleached on exposure to sulphur dioxide,
or the fumes of burning sulphur. This fact, which was known to
Paracelsus, has been explained by two different theories : (i) the
formation of colourless addition compounds with the colouring
matters ; (ii) the reduction of the colours to colourless compounds,
possibly by nascent hydrogen : SO2 -f- 2H2O = H2SO4 -f 2H.

Exri. 186. — Add a few drops of fuchsine ( "magenta") solution
to a solution of sulphur dioxide : the red colour is discharged. Boil
with dilute sulphuric acid : the colour is restored.

EXPT. 187. — To a tincture of red cabbage, prepared by soaking the
dry leaves in alcohol, add sulphurous acid, and neutralise with soda ;
the colour is discharged. If an acid is now added a red colour is formed.
Red roses may be bleached by wetting them, and suspending in a
bell -jar over burning sulphur ; on dipping the flowers into dilute sul-
phuric acid the colour is restored.

Sulphurous acids and sulphites are reducing agents ; they liberate
iodine from potassium iodate :

2KI03.+ 5S02 + 4H20 = I2 + 2KHS04 -f 3H2S04.

The titration of the liberated iodine serves as a means of estimation
of S02 in flue-gases, or sulphites in solution. With excess of
sulphur dioxide, the colour of the iodine again disappears :
I2 + S02 + 2H2O - 2HI 4- H2S04. Titration with iodine may
also be used in the estimation, but the concentration of S02 in the
solution should not exceed 0*04-0 -05 per cent. (Bunsen). The
solution of SO 2 readily absorbs atmospheric oxygen ; the rate of
oxidation is greatly reduced by the addition of glycerin or mannitol.

Titoff (1903) concluded that in perfectly pure water no oxidation would
occur ; oxidation is due to traces of iron and copper salts in all water,
which act as catalysts. Even 1 gm. atom of Cu" in 10° litres exerts
an appreciable influence. Organic substances probably form complex
compounds with the metal ions, their action as negative catalysts there-
fore consists in their capability of destroying the positive catalysts
(Cu", etc.).

Sulphur dioxide decolorises a solution of potassium perman-
ganate :

2KMn04 + 5S02 + 2H2O = K2S04 + 2MnS04 + 2H2S04.

xxvr THE OXYGEN COMPOUNDS OF SULPHUR 105

Sulphites. — Sulphurous acid is dibasic, and forms two series of
salts :

Acid sulphites. Normal sulphites.

KHS03 K2SO3

NaHSOo KNaSO3

Ca(HS03)2 CaSO3

EXPT. 188. — Divide a solution of caustic soda into two equal parts.
Saturate one with SO2, producing a solution of sodium hydrogen sulphite,
NaHSO3. This is acid, owing to dissociation of the HSO3 ion (HSO3'—
SO3" -f- H'). Mix this with the other half of the caustic soda, and
evaporate. Crystals of normal sodium sulphite, Na2SO3,7H2O, are
produced on cooling.

Sodium sulphite forms a slightly alkaline solution, owing to
hydrolysis : SO/ -f H20 — HS03' -f OH'. It gives a white
precipitate of barium sulphite, soluble in hydrochloric acid, on addition
of barium chloride : Ba" + SO/ ^± BaSO3 (dissd.) ^ BaSO3 (ppd.).
If chlorine- or bromine- water is added to the solution .in hydro-
chloric acid, oxidation occurs, and a white precipitate of barium
sulphate, BaS04, insoluble in hydrochloric acid, is formed :

S03" + H20 + C12 = SO/ -f 2C1' + 2H\

Sulphur dioxide, when passed through lime-water, gives a white
precipitate of calcium sulphite, CaS03.

If a solution of sodium hydrogen sulphite is mixed with alcohol,
the salt NaHS03 is precipitated, but if it is boiled and evaporated,
a new salt, Na2S2O5, called sodium metabisulphite (i.e., Na20,2SO2)
is formed, which is used in photography. On heating dry sodium
sulphite, the sulphate and sulphide are formed : 4Na2S03 =
Na2S -f 3Na2S04. NaHS03 on heating first produces Na2S03,
H20, and SO2, and the Na2S03 then decomposes as above.

Thionyl chloride. — If sulphur dioxide is passed over phosphorus
pentachloride, PC15, a liquid is formed which on fractional distilla-
tion is separated into thionyl chloride, SOC12 (b.-pt. 78°), and phos-
phorus oxychloride, POC13 (b.-pt. 107°) : S02 + PC15 =
SOC12 -f POC13. SOC12 is also formed by the addition'of sulphur
to chlorine monoxide at — 12° : C120 + S = SOC12. It is manu-
factured by adding sulphur trioxide to sulphur chloride at 75-80°,
and passing a stream of chlorine through the mixture to reconvert
the separated sulphur into the chloride :

S03 + S2C12 = SOC12 + S02 + S.
Thionyl chloride, i.e., the chloride of the radical thionyl, S0<^ , is

a colourless liquid, sp. gr. 1-675 at 0°. It fumes in moist air, and
is decomposed by water, forming hydrochloric and sulphurous

496 INORGANIC CHEMISTRY CHAP.

acids ; it is an acid chloride i.e., sulphurous acid with univalent
hydroxyl replaced by chlorine :

/a OH

S0 2

\C1 \OH

Thionyl bromide, SOBr2, is a red liquid, b.-pt. 68°/40 mm., formed
by acting on SOC12 with KBr. With SOC12 and HBr a pale yellow
liquid, thionyl chlorobromide, SOClBr, is also formed, b.-pt. 115°.
Thionyl fluoride, SOF2, is a colourless gas obtained by heating SOC12
and arsenic fluoride, AsF3. It boils at — 32°, and forms with dry
ammonia the compounds 2SOF2,5NH3 and 2SOF2,7NH3.

The constitution of sulphurous acid. — The formation of sulphurous
acid by the action of water on thionyl chloride suggests that it has the

XOH

symmetrical formula SO<^ . By the action of thionyl chloride on
\OH

/OC2H6
alcohol symmetrical diethyl sulphite, SO/ , b.-pt. 161°, is formed,

\OC2H5

which is hydrolysed when boiled with caustic soda, yielding alcohol
and sodium sulphite. The formula of the latter would thus appear to be

/ONa
symmetrical : SO<f

xONa

By the action of sodium sulphite, on ethyl iodide, a compound having
the same composition as diethyl sulphite is obtained : Na.>SO3 -f
2C2H?I = (C2H5)2S03 + 2NaI.

This is, however, not symmetrical diethyl sulphite, since it boils at
207°. When boiled with caustic soda it yields sodium ethylsulphonate,
NaC2H5SO3, a salt of ethylsulphonic acid, C2H5SO3H. In the latter the
ethyl group, C2H5, is almost certainly directly attached to the sulphur
atom, since the compound is formed by oxidising mercaptan, or ethyl
hydrogen sulphide, C2H6SH, with dilute nitric acid. The liquid boiling
at 207° also, probably, contains an ethyl group directly attached to the
sulphur atom, since it is derived from the sulphonic acid ; it has an

unsymmetrical formula: ^S\ , whilst the first compound

O^ X0-C,H5
/0-C2H5
is symmetrical : O = S( . These two compounds, which have

X0-C2H5

the same percentage composition, and the same molecular weight, but
different properties, are called meatmeric compounds, or metamers, and the
phenomenon of the existence of such compounds is called metamerism.
Isomerism is explained by the different modes of arrangement of the
atoms in the molecules, i.e., different structures.

xxvi THE OXYGEN COMPOUNDS OF SULPHUR 497

Since sulphurous acid appears to have two different formulae according
to the reactions by which it is produced, viz.,

,H /OH

NDH XOH

unsymmetrical symmetrical

it is assumed that both forms exist in equilibrium in a solution of the acid,
and are readily converted into each other, so that according to the
reagent presented to the acid, the latter appears to have sometimes one
and sometimes the other formula. This property is called dynamic
isomerism, or tautomerism. The two forms are called dynamic isomers.

SULPHUR TRIOXIDE AND SULPHURIC ACID.

Sulphur trioxide. — Sulphur trioxide, S03, is produced by the direct
union of the gaseous dioxide with ozone (Brodie) : 3S02 -f O3 .=
3S03. It is a white crystalline solid. It is also produced when a
mixture of the dioxide and oxygen, or air, is passed over a catalyst,
such as platinised asbestos heated to 500° (P. Phillips, 1831), or the
oxides of iron, copper, chromium, or vanadium heated to 600-700°
(Wohler): 2SO2 -f O2 =2S03. A state of equilibrium is set up,
since the reaction is reversible;

At 400°, 2 per cent, of S03 is decomposed ; at 700°, 40 per cent.
In a mixture of S02 and air, such as is obtained by burning pyrites
(p. 503), containing by volume 7 per cent, of S02, 104 per cent, of
02, and 82-6 per cent, of N2, the following percentages of S02 are
oxidised to S03 in equilibrium : at 434°, 97 ; at 550°, 85 ; at 645°,
60. The reverse change, 2S03 -> 2S02 -f 02, is favoured by rise
of temperature, since it absorbs heat (p. 355). The direct change
2S02 + O2 — 2S03 H- 45 kg. cal. does not proceed in presence of
platinum at an appreciable rate below 400°, on account of the
slowness of reaction at lower temperatures. The two conflicting
effects of temperature on the yield are balanced in practice by work-
ing at 400-450°, which is the optimum temperature with platinum
as a catalyst, and using excess of oxygen in the form of air, as
described.

EXPT. 189. — Pass a mixture of SO2 and O2 through sulphuric acid
to dry it, and then over platinised asbestos heated in a hard glass
tube, c (Fig. 270). Dense white fumes are produced, which condense in
the cooled dry receiver, d, to a colourless liquid, which gradually solidi-
fies. This is sulphur trioxide.

The trioxide is also produced by heating concentrated sulphuric
acid with phosphorus pentoxide : H2S04 -f P2O5 = S03 -f 2HP03,
or most conveniently by distilling fuming sulphuric acid (q.v.) :

K K

498

INORGANIC CHEMISTRY

H2S2O7 ^ H2S04 -f- SO3. If sodium hydrogen sulphate is heated
to 300° it forms the pyrosulphate, and this evolves sulphur trioxide
at a bright red heat : 2NaHSO4 = Na2S2O7 + H2O ; Na2S207 =
Na2SO4 -f S03. The formation of a " volatile salt " on distilling
fuming sulphuric acid was described by " Basil Valentine " (p. 29),
and by Bernhardt in 1775.

Sulphur trioxide appears to exist in two modifications. The liquid,
b.-pt. 46°, at first obtained solidifies on cooling to transparent
crystals, melting at 14-8°, sp. gr. 1-97 at 20°. The lowering of
vapour pressure of a solution of this form, called a-S03, shows that
it has the formula S0a. If a-S03 is allowed to stand some time, in

FIG. 270.— Formation of Sulphur Trioxide.

presence of a trace of moisture, it forms silky,' asbestos-like crystals,
the molecular weight of which, in solution in POC13, corresponds
with S206. This is /?-S03. At 50° the /J-form changes slowly into
the a-form.

The vapour density of sulphur trioxide corresponds with the
formula S03. When passed through a red-hot tube it decomposes,
giving 2 vols. of S02 and 1 vol. of 02, which do not recombine on
cooling except in the presence of a catalyst : 2S03 = 2S02 -f- O2.
The solid absorbs moisture with avidity from the air, giving oft
dense white fumes composed of droplets of sulphuric acid :
H20 + S03 = H2S04. It dissolves in water with a loud hissing
noise, and considerable evolution of heat. Sulphur trioxide dis-

XXVI

THE OXYGEN COMPOUNDS OF SULPHUR

499

solves readily in concentrated sulphuric acid ; the fuming acid
so obtained solidifies on cooling to colourless crystals of pyro-
sulphuric acid, H2S2O7, m.-pt. 35°. Sulphur trioxide reacts vio-
lently with baryta, the mass becoming incandescent • SOo -4- BaO =
BaSO4.

Manufacture of sulphur trioxide and sulphuric acid by the contact
process. — Repeated attempts were made to adapt Phillips's contact
process to large-scale working, but it was found that the platinum
rapidly became inactive (" poisoned "), and the conversion of SO2

FIG. 271 . — Contact Chamber of Badische
Process.

FIG. 272. — Schroder-Grillo Contact
Chamber.

into S03 ceased. After years of unremitting work, the Badische
Soda and Aniline Co., of Ludwigshafen in Germany, found that
the poisoning is due to impurities, especially arsenious oxide
and dust, in the gases from the pyrites burners, and that if these
impurities are got rid of by blowing a jet of steam into the burner
gas, allowing to settle, cooling, and passing through coke filters
drenched with concentrated sulphuric acid until no fog is seen by
a powerful beam of light (" optically clear " gas), the platinum
retains its activity for an indefinite period.

K K 2

500

INORGANIC CHEMISTRY

CHAP.

In the Badische process the purified gas is passed into a converter
(Fig. 271), consisting of an iron cylinder with inlet and outlet pipes.
Inside are vertical iron tubes packed with platinised asbestos. Twice
the theoretical amount of oxygen is present in the gas (in the form
of air), and the apparatus is heated by the gas burners, h, to start
the reaction : this afterwards proceeds automatically. By letting
the incoming gas sweep over the outside of the hot tubes in which
the reaction occurs, no external heating is needed, since a con-
siderable amount of heat is evolved, and the process goes on con-
tinuously at 400-450°.

The sulphur trioxide cannot be absorbed from the issuing gases

FIG. 273.— Mannheim Contact Process.

by passing through water, as a dense fog of minute droplets of
H2S04 is thus formed, which cannot be condensed. The gas is
therefore passed into 97-99 per cent, sulphuric acid in iron towers ;
the concentrated acid rapidly absorbs the SO3, producing fuming
sulphuric acid, or oleum ; or, if a regulated stream of water is
admitted, the 97-99 per cent, acid is continuously increased in
quantity by the reaction S03 + H2O = H2S04 occurring in the
liquid acid/

In the Schroder- Grillo process, which has been extensively worked
in England and America, the catalyst is prepared by moistening
Epsom salt, MgS04,7H2O, with a solution of platinum chloride,
and heating. The salt loses water, and swells up to a voluminous
mass, on which the platinum is very finely divided. This contact

xxvi THE OXYGEN COMPOUNDS OF SULPHUR 501

mass is put on shelves in iron converters, lagged outside (Fig. 272),
and when the process is once started it goes on without external
heating.

The Tenteleff process utilises a catalyst composed of asbestos
" sponge -cloths," soaked in platinic chloride, and the latter re-
duced by formaldehyde. These are ignited, and a number of
superposed cloths are fitted into an iron frame, 3 ft. by 2 ft., inter-
posed in the gas current. The temperature is 450-500°. This
arrangement is often used in finishing off the conversion in the
Mannheim process (q.v.).

The Mannheim process utilises burnt pyrites (Fe2O3 and CuO)
as the contact mass. This is filled into a rectangular tower, the
lower part of which communicates with two pyrites burners, to
which air dried in a sulphuric acid tower is supplied (Fig. 273).
The hot gases pass directly to the iron oxide shaft, and, on account
of the higher temperature, only about 60 per cent, of the S02 is con-
verted into SO3. The arsenious oxide in the burner gases is kept
back in the oxide of iron as ferric arsenate, and after the S03
has been absorbed from the exit gas by sulphuric acid, the gas
is filtered through scrubbers of coke soaked in concentrated sulphuric
acid, reheated, and passed to a Tenteleff converter to finish the
conversion. This process has also been used fairly extensively
in England, but is not so economical as the Schroder method.

Fuming sulphuric acid, or oleum, is an oily liquid, often coloured
brown by organic matter, but colourless when pure, which emits
thick white fumes in moist air. It may be kept in mild-steel drums,
but cracks cast iron (which resists the action of ordinary con-
centrated sulphuric acid). It is made with different contents of
free S03, i.e., S03 in excess of the amount required to form H2S04.
The strongest product contains 60 per cent, of free SO3, and emits
very dense fumes. The hydrates H20,S03 (H2S04, or mono-
hydrate, m.-pt. 10°), H2S04,H20 or S03,2H2O (m.-pt. -8°),
H20,2S03 or H2S207 (pyrosulphuric acid, rn.-pt. 35°), and
H2S04,4H20 (m.-pt. — 25°) are known. Acids containing more
than 60 and less than 40 per cent, of free S03 are liquid at the ordinary
temperature ; the others are solid. Oleum is used in organic
chemistry to prepare sulphonic acids, RS03H (p. 511), and in adding
to mixtures of nitric and sulphuric acids used for nitration (p. 569).
It is added to ordinary acid to increase its concentration.

Sulphuric acid, H2S04. — Moistened floAvers of sulphur, when
exposed to air, are slowly oxidised, and a little sulphuric acid is
formed. Sulphurous acid in solution oxidises slowly when exposed
to air : 2H2S03 -f 02 = 2H2S04. Oxidation occurs more rapidly
when hydrogen peroxide is shaken in a jar of sulphur dioxide :
S02 -f H202 = H2SO4. Chlorine water and bromine water also
oxidise sulphurous acid : H2S03 + H20 + C12 = H2S04 + 2HCL

502 INORGANIC CHEMISTRY CHAP.

Sulphuric acid, or oil of vitriol, is mentioned by the Latin
Geber, who obtained it by distilling green vitriol, i.e., ferrous sul-
phate : 2FeSO4=-FeIO8-f S02-fSO8; with moisture, the SO3
forms H2S04. In 1666 Lemery and Le Fevre obtained the acid
by deflagrating a mixture of sulphur and nitre over a dish of water
under a glass bell, and a small works using this process was estab-
lished in 1740 by Ward, at Richmond. The acid obtained was
called oil of vitriol per campanum. Roebuck, in 1746, replaced
the fragile glass vessels by lead chambers, 6 ft. wide, at Prestonpans,
and these were enlarged in later works. These chambers were
introduced into France in 1769 by the Englishman Holker, and in
1774 La Folie employed a jet of steam in the chamber. A consider-
able advance was possible after the researches of Clement and
Desormes (1793), who pointed out the importance of a current
of air in the chambers, and in 1806 these chemists gave a correct
interpretation of the reactions occurring in the chambers, par-
ticularly the part played by the oxides of nitrogen. A continuous
process, in which the sulphur dioxide was produced from sulphur in
separate burners, and admitted, together with nitrous fumes, air,
and steam, to the chambers, was introduced by Holker into the
French works of Chaptal in 1810. The use of pyrites as a source
of sulphur dioxide, introduced by Hill, of Deptford, in 1818, and
the invention of the Gay-Lussac and Glover towers (q.v.) in 1835
and 1859, respectively, led to the modern chamber acid industry.

More than one million tons of sulphuric acid are produced annually
by each of the three countries, Great Britain, Germany, and
America.

The lead chamber process. — The reactions in the lead chambers
occur between sulphur dioxide, oxygen (air), steam (or water-spray),
and oxides of nitrogen (" nitrous fumes "). It appears, as was
discovered by Clement and Desormes, that an intermediate com-
pound, nitrososulphuric acid (" chamber crystals "), is formed and
decomposed alternately :

(1) 2SO2 + N2O3 + O2 + H2O = 2S02(OH)-ONO

(2) 2SOa(OH)-O-NO + H20 = 2H2SO4 + N203.

The nitrous fumes, N203 (or, really, N02 -f- NO)/ thus act over
and over again in a cyclic manner, i.e., as a catalyst (p. 166).

EXPT. 190. — A dry 6-litre flask, A, is fitted with four inlet tubes, as
shown in Fig. 274, and a small outlet tube. Three of the tubes are con-
nected with wash -bottles containing concentrated sulphuric acid. One of
these is connected with a siphon of liquid SO2, one to a gas-holder con-
taining oxygen, and the third to a gas-holder containing nitric oxide
(p. 578). The fourth tube is connected with a small flask, B, containing
water, which may be heated, and through which oxygen may be bubbled.
.A rapid stream of oxygen is first passed through the apparatus. Nitric

XXVI

THE OXYGEN COMPOUNDS OF SULPHUK

503

oxide is then passed in, which at once forms brown fumes of higher
oxides of nitrogen. Sulphur dioxide is then passed in at the same rate
as the nitric oxide, and, after a short time, a current of oxygen is passed
through the hot water in B to carry moisture into the globe. White
star -shaped crystals of nitrososulphuric acid form on the inside of the
flask. The colour of the gases at the same time becomes much paler.
Sweep out the gases by a rapid current of dry oxygen, and then boil the
water in B. When the steam comes in contact with the crystals,
these dissolve with effervescence, producing red fumes of oxides of
nitrogen.

The liquid in the flask gives a white precipitate (BaSO4) with BaCl2
solution.

On the large scale, lump pyrites is burnt in brick furnaces, called
pyrites burners, the grates of which are composed of separate square
bars which can be
turned on their
longitudinal axes so
as to drop the burnt
ore into the ash-
pits. The supply
of air is carefully
regulated by sliding
doors above and
below the bed of
pyrites. Each fur-
nace holds 3—5 tons
of ore, and

are N0

arranged in sets of
20-25, with a com-
municating flue, for FIG" 274-~EXperimefcid11Chambne?sR^ * SUlphUriC

each set of cham-
bers. The daily charge for each furnace is 750-1000 Ib. of pyrites.
Pyrites powder, or " spent oxide "( p. 477), is burnt in rotary kilns,
consisting of iron cylinders lined with firebrick, with a series of
shelves so arranged that the ore is raked from shelf to shelf until
the burnt ore is discharged at the bottom. The rakes are actuated
by a revolving air- or water-cooled central shaft.

The burner gas (7 per cent, of S02, 10 per cent, of 02, 83 per
cent, of N2) passes to a dust-catcher, containing baffle-walls, and then
through a nitre-oven, in which pots containing sodium nitrate and
sulphuric acid are placed. These supply the oxides of nitrogen to
make up losses from the plant. About 3 parts of NaNO3 per 100
parts of S burnt as pyrites are required. In modern plants, the
oxides of nitrogen are supplied by the oxidation of ammonia (p. 575),
From the dust-catcher the gases pass, at 300-400°, into the Glovei

504 INORGANIC CHEMISTRY CHAP.

tower, seen on the right in Fig. 275. This is a lead tower lined with
acid-resisting bricks, 20-30 ft. high, and 6-8 ft. diameter, packed
with flints resting on an arch. Down this tower two streams of
acid, from the tanks seen at the top, are passed. One stream con-
sists of dilute acid (65-70 per cent. H2SO4) from the lead chambers ;
the other consists of stronger acid (78 per cent. H2SO4) containing
oxides of nitrogen (in the form of nitrososulphuric acid) from the
Gay-Lussac tower seen on the left. The functions of the Glover
tower are three : (a] to cool the burner gases to 50-80° before
they enter the chambers ; (b) to denitrate the acid from the Gay-

FlG. 275. — Diagram of Sulphuric Acid Chamber Plant, showing end view of three Chambers,
Gay-Lussac Tower (left), Glover Tower (right) and Pyrites Burners, A.

Lussac tower, by dilution with chamber acid, and heating ; (c) to
concentrate the weak acid to about 78 per cent. H2S04 for sale,
or for use in the Gay-Lussac tower, and at the same time provide
steam for the chambers. About 25 per cent, of the acid made in
the plant is also formed by reactions in the Glover tower.

From the Glover tower the gases now pass, by a lead main seen
on the extreme right of Fig. 275, to the first of the set of lead
chambers, the ends of three of which are shown. These are formed
of sheet lead weighing 6—8 Ib. per sq. ft., are oblong or square in
shape, and dip into large lead saucers with a seal of acid. The
chambers are suspended from a wooden or iron frame by lead straps

xxvi THE OXYGEN COMPOUNDS OF SULPHUR f>(>5

welded on the sides. All joints in the lead sheets are autogenously
welded by a hydrogen flame. The capacity of each chamber is
25.000-75,000 cu. ft., and three or four, connected by wide lead
pipes, form a set.

Steam, or more usually a fine spray of liquid water from several
jets in the roof, is blown into the chamber. Sulphuric acid is
produced in the form of a fog of small drops, which settle down into
liquid chamber acid (65-70 per cent. H2S04) on the floor of the
chamber. In modern practice, 10 cu. ft. of chamber space is
allowed per Ib. of S burnt per twenty -four hours. The capacities
of the Glover and Gay-Lussac towers are each about 1 per cent,
that of the chambers ; the height of the Glover tower does not
exceed 30 ft. The conversion of S02 to H2S04 reaches 98 per cent.

The gases from the last chamber, containing nitrogen, a little
oxygen, most of the oxides of nitrogen in circulation through the
plant, and a trace of sulphur dioxide, now pass to the Gay-Lussac
tower, shown on the left in Fig. 275. This is a lined lead tower,
40-60 ft. high, and 8-15 ft. in diameter, packed with hard coke, and
fed with cold Glover acid (78 per cent. H?S04). Its function is to
recover the oxides of nitrogen in the exit gases from the chambers.
These are absorbed, producing nitrous vitriol, containing nitroso-
sulphuric acid equivalent to 1-2 per cent. N203, which is pumped
to the Glover tower for denitration. The waste gas from the Gay-
Lussac tower passes to a chimney, which maintains a draught
through the whole system.

Theory of the lead chamber process. — The reactions which occur
in the chambers are complicated, and still not completely under-
stood. The chief point calling for explanation is the action of the
oxides of nitrogen. Berzelius represented this as follows :

(1) NO2 -f S02 + H20 = H2SO4 + NO.

(2) NO -f O = N02.

Davy put forward another explanation, elaborated by Lunge.
According to this, nitrososulphuric acid, SO2(OH)-0-NO, i.e.,
sulphuric acid, S02(OH)-OH, in which one atom of hydrogen is
replaced by the nitroso-group, NO, is an intermediate product.
This is formed by the action of nitrous anhydride, N203, traces of
which exist in equilibrium with NO and N02 ': NO + N02 =; N203 :

(1) 2S02 + N203 + 08 + H20 = 2S02(OH)-0-NO.
This does not deposit in crystals, but is at once hydrolysed :

(2) 2S02(OH)-ONO + H20 = 2S02(OH)2 + N2O3.
In the first chamber, where the gases are very pale and an excess
of NO is present, the following reactions may occur :

2S02(OH)-0-NO + S02 + 2H2O = 3S02(OH)2 + 2NO
2S02 + 2NO + 30 + H20 = 2SO2(OH)-0-NO.

506 INORGANIC CHEMISTRY . CHAP.

In a more recent theory Lunge (1906) assumes the formation of a
hypothetical sulphonitronic acid, H2SNO5, which then forms nitroso-
sulphuric acid, HSNO5 :

(1) S02 + N02 + H20 = H2SN05.

(2a) 2H2SN05 + O - H2O + 2HSNO5.

(26) 2H2SN05 + N02 = 2HSN05 + NO + H20.

Decomposition of nitrososulphuric acid then occurs :

(3ft) 2HSNO5 + H20 = 2H2SO4 + NO + N02.
(36) 2HSN05 + S02 + 2H2O = H2S04 + 2H2SN05.
(3c) H2SN05 =-- NO + H2S04.
(4) NO + O = N02.

Raschig (1887) proposed a different scheme :

/OH
(1) 2HN02 + S02 = 0 : N< + NO

(2) H2SN05 = H2S04 + NO.

(3) 2NO + H20 + O = 2HN02.

He further supposes that H2SN05 may react with SO2 to form
hydroxylamine disulphonic acid, HO-N(S03H)2, and nitrylsulphonic acid,
N(SO3H)3, which may lose sulphonic groups and form hydroxyl-
aminc, NH2OH, and ammonia, NH3. Traces of the latter are some-
times found in chamber acid. Trautz believes that, to a limited
extent, nitrosodisulphonic acid, NO(S03H)2, may be formed by the
action of nitrous acid on sulphurous acid, which is then decomposed
by nitrous acid as follows :

NO(S03H)2 + 2HO-NO = 3NO + 2H2S04.

The concentration of sulphuric acid. — The chamber acid (65-70
per cent. H2SO4) may be used directly in the manufacture of super-
phosphates. Unless all the acid is passed through the Glover tower,
the remainder of the chamber acid may be concentrated to the
strength of Glover tower acid (78 per cent. H2S04) by evaporation
in flat lead pans by waste heat from the pyrites burners. The 78 per
cent, acid is usually called " brown oil of vitriol," or B.O.V., on
account of its colour, due to organic matter. Stronger acid, 93—95
per cent. H2S04, called " rectified oil of vitriol," or R.O.V., is
required for many purposes, and is made by concentration
of B.O.V. This concentration, formerly carried out by heating in
glass or platinum retorts, when steam is emitted, is now effected
in one of three types of concentration apparatus : the Cascade
apparatus, the Kessler apparatus, and the Gaillard tower. In all
cases the acid is heated and a current of hot air passed over its
surface. The vapours emitted are composed of very weak acid
so that the remaining acid increases in strength.

XXVI

THE OXYGEN COMPOUNDS OF SULPHUR

507

In the cascade process the acid is allowed to flow down a series of
vitrified silica, or ferro- silicon, dishes, arranged one above the
other, with the spout of one discharging into the basin next lower,
the whole resting on a kind of staircase of acid-resisting bricks. The
acid is heated by a flue below, and hot air sweeps over its surface
(Fig. 276). Much of the acid is lost in the form of fumes. From the
last dish, which may be of cast iron, the acid flows to a cooler.

In the Kessler apparatus the acid flows through a dish, S, of Volvic
stone (a natural acid-resisting material of volcanic origin, found at
Puy-de-D6me), covered outside with lead, through which hot gas
from a coke generator (p. 705) passes (Fig. 277). The dish has
ridges, b, so as to bring the acid and fire-gas into intimate contact.
The concentrated acid runs off to a cooler. The fumes pass through
a tower, R, containing a number of plates with perforations covered
with inverted cups, down which the acid to be concentrated is fed.

FIG. 276. — Cascade Apparatus for Concentrating Sulphuric Acid.

Much of the fume is here condensed, and the temperature is kept
at such a point that steam escapes, but the sulphuric acid remains.
The issuing fumes then pass through a lead box packed with graded
coke, drenched with concentrated sulphuric acid, which takes out
the fine mist of acid droplets.

The Gaillard tower (Fig. 278) consists of an empty tower of Volvic
stone or acid-resisting brick, from the top of which a fine spray of
acid is discharged. In passing down the tower this spray meets a
current of hot gas from a coke generator, which enters the tower
at the side near the bottom. The acid is concentrated by the hot
gas, and runs out from the lead saucer, in which the tower stands,
to a cooler. The fumes from the tower are passed through a smaller
empty lead tower, called a recuperator, down which a portion of the
acid to be concentrated is passed, and finally to coke filters. The
tower is 60 ft. high and 10 ft. in diameter.

508

INORGANIC CHEMISTRY

The acid fumes from concentrators may be condensed by means of
electrostatic precipitation (p. 15). They are passed through a chamber
in which lead plates are hung, with lead covered bars hanging vertically
between them. These are charged to a potential of 20,000 volts. The
acid droplets are attracted to the plates, and the liquid deposited on the
latter runs off to collecting tanks, and is returned to the concentrators.

Still more concentrated acid, containing 97-98 per cent, of H2S04,
is produced from the 93-95 per cent, acid by heating the latter in

cast-iron pans by direct fire. The
very strongest acid does not
attack cast iron, whilst 93-95
per cent, acid dissolves it. The
acid is therefore run in a fine
stream on to the surface of a
large bulk of 98 per cent, acid
boiling in a large cast-iron
pot provided with a siphon
neck opening into it near the
bottom. The strong acid is
run off continuously from this
" swan -neck " as the concen-
tration proceeds. The acid
may also be brought to any
desired strength by the addi-
tion of oleum (sulphuric acid
containing S03).

The purification of sulphuric
acid. — Commercial sulphuric
acid often contains arsenic
trioxide, As203, in solution,
derived from the arsenic in
the pyrites. It is purified by
treating the chamber acid with
sulphuretted hydrogen in lead
towers or closed agitators.
The precipitate of arsenic sul-
phide, As2S3, is filtered off
by suction through unglazed
earthenware plates, or is re-
moved by flotation (p. 10) ;
a little paraffin, added to the liquid, floats to the surface
and carries with it the precipitate. Acid made from sulphur
(" brimstone acid ") is preferred for the preparation of foods (e.g.,
glucose, made from starch by boiling with dilute sulphuric acid),
although de-arsenicated acid from pyrites is also used.

THE OXYGEN COMPOUNDS OF SULPHUR

509

Properties of sulphuric acid. — Pure sulphuric acid, or monohydrate,
H2S04, is prepared by adding the requisite amount of S03 to
98 per cent. acid. It is an oily liquid which fumes slightly in air,
apparently because it contains a little free sulphur trioxide :
H2S04^S03 -f H2O, formed by dissociation in the liquid. This
dissociation increases on heating, and the vapour is richer in SO3
than the residual liquid. It is therefore impossible to obtain pure
H2S04 by the ordinary concentration process. The monohydrate
freezes in ice and salt, and the crystals then melt at 10°. On
boiling, an acid of constant composition, 98-3 per cent. H2S04,
comes over at a temperature of 330°, which is usually given as the
boiling point of sulphuric acid.

The ordinary concentrated acid, containing about 98 per cent.

FIG. 278. — Gaillard Tower for Concentrating Sulphuric Acid.

H2S04, is an oily colourless liquid, of sp. gr. 1-85. It does not
fume in the air.

Concentrated sulphuric acid has a strong affinity for water. When
the acid is mixed with water a considerable amount of heat is given
out, and the liquid may boil. In practice, it is always safest to add
the acid to the water in a thin stream, with stirring. The water
should never be added to the acid. The diluted acid occupies a
less volume than its constituents.

If the acid is mixed with snow, cold is produced, because the
latent heat of fusion of ice exceeds the heat evolved on mixing the
acid with liquid water.

INORGANIC CHEMISTRY

CHAP.

The definite crystalline hydrates, H2SO4,H2O and H2SO4,4H2O, are
known, and probably exist in a partially dissociated state in the liquid.

The density of pure sulphuric acid is 1-8384 at 15°. The densities of
mixtures of the acid with water are given in the table below. It will be
seen that 97-7 per cent, acid has a maximum density.

TABLE or DENSITIES OF SULPHURIC ACID.

Density.
1-000
1-075
1-150
1-180
1-240
1-300
1-390
1-450
1-525
1-600

Density.
1-888
1-920
1-957
1-979
2-009

Per cent,
H2S04.
0-09
10-90
20-91
24-76
32-28
39-19
49-06
55-03
62-06
68-51

Density.
1-645
1-720
1-780
1-825
1-835
1-841
1-8415
1-840
1-8384

TABLE OF DENSITIES OF OLEUM.

Per cent.

free SO3. Density.

10 2-020

20 2-018

30 2-008

40 1-990
50 1-984

Per cent.
H2S04.
72-40
78-92
84-50
91-00
93-43
97-00
97-70
99-20

100.00

Per cent,
free SO3.

90
100

These tables show that, at higher strengths, the density does not
enable one to find the concentration of the acid.

On account of its great affinity for water, concentrated sulphuric
acid is used for drying gases on which it does not act chemically.
It is most conveniently spread over pumice by boiling pieces of
this substance with the acid ; the lumps of impregnated pumice
are placed in a glass tower.

The affinity of strong sulphuric acid for water is also shown by
the charring of organic matter containing carbon, hydrogen, and
oxygen, by the acid. The elements of water are removed, and black
carbon is left.

EXPT. 191. — To a strong syrup of cane-sugar, C12H22OU, contained in
a beaker standing in a stoneware trough, add concentrated sulphuric
acid. The mixture rapidly becomes dark in colour and hot, and froths
up into a black mass of finely-divided carbon, clouds of steam and sulphur
dioxide being evolved. If the black mass is washed with water on a
filter-paper, a dark brown colloidal solution of carbon passes through.

XXVI

THE OXYGEN COMPOUNDS OF SULPHUR

511

Other organic substances, such as wood, are charred ; pure
cellulose, such as filter-paper, is not charred by the cold, slightly
diluted, acid, but forms a tough, semi-transparent material, known
as parchment-paper, which since it is impervious to fats is used for
wrapping butter and other greasy materials.

By heating concentrated sulphuric acid with benzene, C6H6,
elimination of water occurs, and benzenesulphonic acid, C6H5'S03H,
is produced : C6H« 4- H2S04 = C6H5'S03H 4- H20. On fusing the
sodium salt of this with caustic soda, the sodium compound of
phenol or " carbolic acid," C6H5OH, is produced :

Na2SO3.

CeH6S08Na

NaOH = C6H5ONa

The sodium phenoxide, C6H5ONa, may be decomposed by an acid
(even carbon dioxide under pressure), and phenol is formed. Many

FIG. 279.— Decomposition of Sulphuric Acid by Heat.

other sulphonic acids, all containing the group S03H— , are pre-
pared and used as intermediate products in the manufacture of dyes,
drugs, etc. Very often fuming sulphuric acid is used in sulphonation.

The vapour density of sulphuric acid at 444° is 25, whilst the
calculated density for complete dissociation into S03 and H20
is (18 + 80)/4 = 24-5. The products recombine on cooling :
H2S04 = H2O + S03. If the vapour is passed through a red-hot
tube of platinum or quartz, the sulphur trioxide is decomposed,
oxygen and sulphur dioxide being produced : 2H2SO4 —
2S02 + 02 + H20.

EXPT. 192. Fit a dropping funnel by means of a mixture of asbestos

powder and thick water-glass (sodium silicate) into a silica tube contain-
ing broken pumice, and connected with a U-tube as shown in Fig. 279.
Heat the tube to bright redness by means of powerful Bunsen burners,

512 INORGANIC CHEMISTRY CHAP.

and allow concentrated sulphuric acid to drop slo\\-]y into it. Any
undecomposed acid collects iii the U-tube, whilst oxygen may bo col-
lected in a gas jar over water.

In aqueous solution sulphuric acid behaves as a strong acid, since
it is largely ionised. The ionisation occurs in two stages, the second
being appreciable only at high dilution :

H9S04 — H' + HS04'
HS04 — IT + SO/.

Two series of sulphates are therefore known, the acid and normal
salts, corresponding with the formulae RHSO4 and R2SO4. Many
of these sulphates are important minerals : gypsum, CaSO4,2H2O ;
anhydrite, CaS04 ; barytes, BaS04, celestine, SrSO4 ; glauberite,
CaS04,NagSO4 ; and kieserite, MgSO4,H2O.

Most sulphates are crystalline, and soluble in water. The sul-
phates of lead, calcium, and strontium are sparingly soluble in
water ; barium sulphate is practically insoluble in water and dilute
acids, and its formation is used as a test for sulphuric acid or soluble
sulphates. A solution of barium chloride is added to the liquid to
be tested, and then dilute hydrochloric acid. The formation of a
white precipitate, BaS04, indicates the presence of the ion, S04".

Care should be taken not to add an excess of concentrated hydro-
chloric acid, as in that case a white precipitate of barium chloride is
thrown down,, on account of the action of the chloride ion (p. 358).
This, however, readily dissolves in water. In the estimation of sulphuric
acid or sulphates, the boiling solution is mixed with boiling solution of
barium chloride. The precipitated BaSO4 is then readily filtered.

Potassium sulphates. — If dilute sulphuric acid is neutralised with
caustic potash, or potassium carbonate, and the solution evaporated,
anhydrous rhombic prisms of potassium sulphate, K2SO4, separate.
These are not very soluble in water (10-3 gm. in 100 gm. of water at
15° ; 24-1 gm. at 100°) ; the solubility increasing almost linearly
with the temperature (Fig. 68). Potassium sulphate melts at
1050°. The salt occurs in large quantities in the double salts of the
Stassfurt potash deposits : schonite, K2S04,MgS04,6H20 ; and
kainite, K2S04,MgS04,MgCl2,6H20.

If kainite is dissolved in hot water, it breaks up into its constituent
salts, which are largely ionised in solution, yielding the ions K',Mg'!,
SO/jCF. By fractional crystallisation, those salts separate first
(including double salts) with which the solution first becomes
saturated (Van't Hoff). From warm solutions the double salt
schonite first separates, since it is least soluble, and magnesium
chloride remains in solution. If the schonite is digested with potass-
ium chloride (occurring at Stassfurt as sylvine], the following

xxvi THE OXYGEN COMPOUNDS OF SULPHUR 513

reaction occurs : K2S04,MgSO4,C>H2O 4 2KC1 ^= 2K2SO4 -f
MgCl2 -f- 6H2O. The potassium sulphate, being sparingly "soluble,
separates first, followed by carnallite, KCl,MgCl2,6H2O, from which
KC1 and MgCl2 can be prepared (p. 791).

Potassium sulphate is also obtained in smaller amounts by the
action of concentrated sulphuric acid on the chloride : 2KC1 4-
H2SO4 = K2S04 + 2HC1 ; and as a by-product in the manufacture
of potassium dichromate (p. 947) and permanganate (p. 966).
Potassium sulphate is used in the preparation of potash alum (p. 899)
and as a fertiliser (p. 789).

If potassium sulphate is heated with an equivalent of concentrated
sulphuric acid, it dissolves ; potassium hydrogen sulphate (" potassium
bisulphate," K20,2S03, or " acid potassium sulphate "), KHS04,
being formed, which fuses at 197° (Roulle, 1754). This is obtained
as a by-product in the preparation of nitric acid (p. 566). It is
readily soluble in water ; the solution giving a strongly acid reaction,
owing to the formation of hydrogen ions :

K- + HSO/
HS04' — H- + S04".

On evaporation, this solution, in accordance with Van't Hoff's
rule, deposits the normal sulphate, K2S04, which is the salt with
which the solution first becomes saturated. The residual solution
contains free sulphuric acid. From this, on cooling, a trisulphate,
K2S04,KHS04, or K20,3SO3,H2O, deposits, and finally KHSO4.
The compounds K2S04,3KHSO4 and K2SO4,6KHS04 are known.

At a red heat, potassium hydrogen sulphate loses water and forms
potassium pyrosulphate : 2KHS04 = H2O 4- K2S207. At higher tem-
peratures this evolves sulphur trioxide : K2S207 = K2S04 + S03 ;
hence it is used to attack refractory minerals in analysis, since it
behaves like sulphuric acid of high boiling point. Thus chromite,
FeO,Or2O3, is converted into ferrous and chromic sulphates, FeSO4
and Cra(SO4)8, although it is not attacked by boiling sulphuric
acid.

Sodium sulphates. — Normal sodium sulphate, Na2S04, is prepared in
large quantities as salt-cake in the first part of the Leblanc process
(p. 777). It crystallises from water as Glauber's salt, Na2S04,10H2O,
forming large monoclinic prisms, which effloresce readily in the air,
and fall to a white powder of anhydrous salt : Na2SO4,10H20 —
Na2SO4 4- 10H2O (vap.). The crystals melt at 32-48°, but deposi-
tion of anhydrous salt simultaneously occurs. The solubility of
Glauber's salt is peculiar, since it reaches a maximum at 32-48°
(Fig. 68). At this temperature the solid in contact with the
solution is converted into the anhydrous salt, the solubility of
which diminishes with further rise of temperature. The solubility
curve therefore consists of two parts, meeting in a sharp angle at

L L

514 INORGANIC CHEMISTRY CHAP.

32-48°, the first part being the solubility curve of Glauber's salt,
and the second part that of anhydrous sodium sulphate.

Glauber's salt readily shows the phenomenon of supersaturation
(p. 101). If the supersaturated solution is brought in contact with a
minute crystal of Glauber's salt, such as one of those which are always
floating in dusty air, crystallisation at once begins, and Glauber's salt is
deposited. But if it is cooled to 5°, it deposits crystals of a metastable
heptahydrate, Na2SO4,7H2O, which become opaque when touched with
a crystal of Glauber's salt, owing to decomposition :

2Na2SO4,7H2O = Na2SO4,10H2O + Na2SO4 + 4H2O.

The anhydrous sulphate occurs as thenardile ; glaufoerite is the
double saltfCaS04,Na2SO4.

Sodium hydrogen sulphate, NaHS04 (" sodium bisulphate "), is
formed in large triclinic prisms by the action of warm concentrated
sulphuric acid on anhydrous sodium sulphate. It is formed in
the preparation of hydrochloric acid (p. 229). A fused mixture,
or compound, of this salt and the normal sulphate is formed as a
by-product in the manufacture of nitric acid (p. 573), and is known
as nitre-cake. The salts NaHS04,H20 and Na2SO4,NaHS04 are
known. The acid sulphate of sodium is decomposed by alcohol into
the salt NaHSOi5Na2SO4, and free sulphuric acid :'3NaHS04 =±
Na2SO4,NaHS04 -f H2S04. Dry KHS04 is not decomposed by
dry alcohol. Sodium pyrosulphate, Na2S207, is formed on gentle
ignition of the acid sulphate, or by the action of sulphur trioxide
on common salt : 2NaCl -f 3S03 = Na2S2O7 4- SO2C12. On heating
to redness, it decomposes into sulphur trioxide and the normal
sulphate. The solution of sodium hydrogen sulphate is acid, for
the same reason as that of the potassium salt, but on evaporation
above 50° it yields crystals of NaHS04.

The chlorides of sulphuric acid. — If sulphuric acid is treated with
phosphorus pentachloride, PC15, hydrogen chloride is evolved, and
two compounds are formed which have the formulae S03HC1 and
S02C12. The reaction involves the replacement of one or two OH
groups, respectively, by Cl, and is similar to the action of the
phosphorus halides on water (p. 640). Since it has been found that
this reaction always occurs when hydroxyl groups are present in a
compound, it is assumed that sulphuric acid has the formula

S02(OH)2, the radical S02 <^ being called sulphuryl. The inter

action of the phosphorus pentachloride is then represented by the
equations :

S09(OH)2 + PC15 - SO2(OH)C1 + POC13 + HC1
SO"2(OH)C1 + PC15 = S02C12 + POC13 + HCl.

The three substances may be separated by fractional distillation,

xxvi THE OXYGEN COMPOUNDS OF SULPHUR 515

since their boiling points are quite different; POOL, 107-2° •
SO2(OH)C1, 155-3° ; S02012, 69°.

The compounds SO2(OH)C1 and S02C12 are known as chlorides of
sulphuric acid ; they belong to the general class of acid chlorides,
which are formed by the exchange of hydroxyl groups for chlorine,
and with water are reconverted into the original acids :

SOa(OH)Cl + H2O = S02(OH)2 + HC1
SO3C12 + 2H2O = SO2(OH)2 + 2HC1.

The two compounds SO2(OH)C1 and SO2C12, known as chlorosul-
phonic acid and sulphuryl chloride, respectively, both contain the
bivalent radical sulphuryl ; the former, if written as S03HC1, is
seen to contain the characteristic grouping, S03H, of sulphonic
acids, hence its name.

Since sulphuric acid contains two hydroxyl groups, the radical
S02 must be bivalent ; the two oxygen atoms of this radical are
united by two valencies each to the sulphur atom, and the latter
must therefore be sexivalent. The graphic formula of sulphuric
acid is therefore :

( OH

Chlorosulphonic acid, S03HC1, may be obtained by the direct
combination of sulphur trioxide and hydrogen chloride : S03 -f-
HC1 = S03HC1, or by the action of phosphorus pentachloride on
sulphuric acid as explained above. Since an excess of the phos-
phorus pentachloride produces sulphuryl chloride, phosphorus
oxychloride, POC13, may be used instead, as this does not interact
further with chlorosulphonic acid : 2S02(OH)2 + POC13 =
2S02(OH)C1 + HP03 + HC1. It is obtained on the large scale by
passing dry hydrogen chloride through fuming sulphuric acid
(containing S03), and distilling. Chlorosulphonic acid is a
colourless, fuming liquid, sp. gr. 1-776, which is violently de-
composed by water, producing sulphuric and hydrochloric acids.
When heated to 170-190° it decomposes into SO2C12 and H2S04 ;
at higher temperatures it breaks down into C12; SO2, and H2O. It
reacts violently with silver nitrate, forming nitrososulphuric acid :

2S03HC1 + 2AgN03 = 2AgCl + 2SO2(OH)ONO + 02.

Sulphuryl chloride, S02C12, is produced by the direct combination
of chlorine and sulphur dioxide in presence of sunlight, or under the
catalytic influence of camphor, glacial acetic acid , or animal charcoal :
S02 -f- Cla ^± S02C12. It is formed by the prolonged action of
phosphorus pentachloride on sulphuric acid, or by heating chloro-
sulphonic acid in a sealed tube at 180°. It may be produced by a
modification of the last reaction, by heating a mixture of chloro-
sulphonic acid with 1 per cent, of mercuric sulphate, which acts as

L L 2

51(5 INORGANIC CHEMISTRY CIJAP.

a catalyst, in a flask under a reflux condenser heated to 70°, and
condensing the vapour.

Sulphuryl chloride is a colourless, fuming liquid which boils at
69° without decomposition. It is rapidly decomposed by water,
with formation of sulphuric and hydrochloric acids ; chlorosul-
phonic acid is formed as an intermediate stage. With ice-cold
water it forms a crystalline hydrate, S02C12,15H20. The direct
formation of SO2C12 from S02 and C12 shows that its graphic formula
contains a sexivalent sulphur atom, since chlorine always adds on
to sulphur in preference to oxygen :

.0 cf ov ,ci

< + i X

X0 Cl CK \C1

The chloride of pyrosulphuric acid, pyrosulphuryl chloride,
S205C12, is obtained by the action of sulphur trioxide on sulphuryl
chloride or on sulphur chloride :

S02C12 + S03 = S205C12

5S03 + S2C12 = 5S02 + S205C12.

It is also formed by the action of sulphur trioxide or chlorosul-
phonic acid on phosphorus pentachloride :

2S03 + PC15 = POC13 -f S205C12

2S02(OH)C1 '+ PC15 = POC13 + 2HC1 + S205C12.

It is a heavy, mobile liquid, sp. gr. 1-844/18°, boiling at 150-7°
under 730 mm. pressure, giving a normal vapour density. It fumes
only slightly and is decomposed only slowly by water": S205C12 4-
3H2O = 2H2S04 + 2HC1. It may be regarded as produced from
2 molecules of chlorosulphonic acid by elimination of water :

/Jl
SO / yd

\iOHi S«< S02

In the same way, pyrosulphuric acid, H2S207, may be regarded
as formed from 2 molecules of sulphuric acid by the elimination of a

molecule of water :

/OH
SO/ /OH

NOH: so/

hm ~>
/° M! so

QPk / ------ ^^2

S°2\OH

xxvi THE OXYGEN COMPOUNDS OF SULPHUR 517

Such reactions, in which certain atoms are removed from two or
more molecules, and the residues combine to form a single molecule,
are called condensations.

The compounds S02(OH)F (b.-pt. 162'6°), obtained by heating
fluorspar with fuming sulphuric acid, S02F2 (b.-pt. — 52°), and
solid S2p3Cl4(S03HCl + SC14 = S2O3C14 + HC1), are known.

Negative groups. — Although all acids contain hydrogen which can
be ionised in solution, there are numerous hydrogen compounds,
such as NH3 and NaH, which have no acidic properties. One
atom of hydrogen in ammonia, NH3, can be replaced by the metals
sodium, potassium, or lithium, forming sodamide, NaNH2, etc.
Hydrogen atoms in hydrocarbons may also, by indirect means, be
replaced by metals, forming organo-metallic compounds ; thus, from
ethane, C2H6, we can obtain zinc ethyl, Zn(C2H5)2. It is therefore
not sufficient that a substance shall contain hydrogen which can be
replaced by metals in order that it shall be an acid. Acidic hydrogen,
however, is always replaced, appearing in the form of water, by
metals presented to it in the form of hydroxides, and this statement
is equivalent to saying that acidic hydrogen is that which can form
hydrogen ions, the latter uniting with hydroxyl to form water
(p. 294) : H' + OH' = H20.

The acidic character of certain hydrogen compounds is determined
by the character of the rest of the molecule of these compounds. In
the hydracids of halogens, for instance, the hydrogen is united with
a strongly electronegative atom of halogen, and strong acids result.
In H2S the hydrogen is united with the weakly electronegative
atom of sulphur, and H2S is a very weak acid. The case of water,
H2O, is exceptional, since it combines both acid and basic functions,
ionising into H' and QH'.

In the oxy-acids, the acidic hydrogen is directly linked to oxygen
as hydroxyl, OH, which, of course, is not usually ionisable as the
hydroxide ion. Thus, the action of water on sulphuryl chloride
gives sulphuric acid, showing that Cl in the S02C12 is replaced
by OH :

/Cl /OH OH

S02< + 2H20 = S02< + 2HC1. The molecule SO
X

C1 OH OH

may therefore be regarded as formed by the replacement of 2 atoms
of hydrogen from 2 molecules of water by the bivalent negative
radical sulphuryl : = S02. This constitution, first deduced by
Williamson (1852), is expressed by saying that sulphuric acid and
other oxy-acids are built up on the water-type.

The acidic character of the hydrogen in oxy-acids is therefore

due to the presence of a negative group, e.g., /S02, in the molecule.

518 INORGANIC CHEMISTRY CHAP.

In organic acids this negative group is uniformly the carbonyl group,
CO. Thus, acetic acid is CEL-CO-OH, and oxalic acid is

(COOH)2.

If hydroxyl is combined with a positive group, such as an atom of
metal, or a radical such as ammonium, NH4, it ionises as such, and
the compound shows basic properties. The more strongly electro-
positive is the metal, or radical, the stronger is the base. Thus,
KOH is a strong base, Fe(OH)3 is a weak base.

If the positive group is only weakly electropositive, the compound
may show weakly acidic properties. Thus, A1(OH)3 behaves either
as a weak base or as a weak acid, according as it is treated with a
strong acid or a strong base (p. 360) : A1(OH)3 + 3HC1 = A1C13 +
3H20 ; Al(OH), + KOH = KA1O2 -f 2H2O. The organic amino-
acetic acid, containing both the positive amino-group, — NH2, and

the negative carbonyl group /CO, is at the same time a weak base

and a weak acid : CH2-NH2-CO-OH. Such a substance is called
amphoteric ; the acidic and basic properties are then very weak and
practically evenly balanced.

PERSULPHUEIC ACIDS.

Persulphuric acids. — Faraday (1832), when electrolysing an
aqueous solution of sulphuric acid, observed that, if the acid were
concentrated; " a remarkable disappearance of oxygen took place."
In 1878 Berthelot exposed a mixture of sulphur dioxide and oxygen
to the silent discharge, and obtained a contraction corresponding
with the formation of S2O7. A small quantity of viscous liquid
separated on the walls of the ozoniser, which solidified at 0° to long
prismatic crystals. This was supposed by Berthelot to be persul-
phuric anhydride, S207. Marshall (1891) found that if a concentrated
solution of potassium hydrogen sulphate, KHS04, is electrolysed,
crystals of the composition KS04 separate at the anode.

In Faraday's experiment persulphuric acid is formed, probably
from the ions HS04' discharged at the anode : H2SO4 ^± H' + HSO4' ;
2HSO4 — H2S208. The doubled formula is confirmed by the deter-
mination of the molecular weight of the potassium, salt by the
freezing-point method ; this is found to be K2S208.

EXPT. 193. — Persulphuric acid is readily formed by the electrolysis
of 50 per cent, sulphuric acid with an anode formed of a fine platinum
point, surrounded by a glass tube to serve as a diaphragm. The cathode
consists of a ring of platinum wire placed outside the diaphragm
(Fig. 280) . The apparatus is kept cool by immersion in a freezing mixture.

XXVI

THE OXYGEN COMPOUNDS OF SULPHUR

519

If potassium hydrogen sulphate solution is used in the same apparatus,
crystals of the persulphate separate out. As strong a solution as pos-
sible should be used. The solution in each experiment gives a brown
colour with potassium iodide : H2S2O8 + 2KI = 2KHSO4 + I2.

In the preparation of potassium persulphate, the ions, HS04',
crowding together at the anode are discharged, and persulphuric
acid is formed : 2HSO4 = H2S2O8. This reacts with the potassium
hydrogen sulphate, and the sparingly soluble persulphate crystal-
lises out : H2S208 + 2KHS04 =± K2S208 + 2H2S04.

A solution of a persulphate acts as a powerful oxidising agent.
Besides slowly liberating iodine from iodides, it oxidises manganous
salts to manganese dioxide, precipitates red copper peroxide, Cu02,
from solutions of copper salts, and black silver peroxide from silver
nitrate. The ammonium salt, (NH4)2S2O8, prepared in the same
way as the potassium salt, is the most soluble persulphate ; it is
used for bleaching, and in photography to " reduce " the intensity of

FIG. 280.— Preparation of Persulphuric Acid.

negatives. The barium salt is very soluble in water, and serves to
separate persulphuric acid from sulphuric acid.

Caro in 1898, by dissolving potassium persulphate in concentrated
sulphuric acid, obtained a solution of a new persulphuric acid,
which was a powerful oxidising agent, converting aniline into nitro-
benzene, but differing from Marshall's acid. This acid, known as
Caro's acid, was investigated by Baeyer and Villiger in 1901. They
prepared it by grinding K2S2O8 with concentrated sulphuric acid,
allowing to stand one hour, and pouring on to ice. Sulphuric acid
was removed by shaking with the sparingly soluble barium phos-
phate. The solution might contain Marshall's acid, Caro's acid,

520 INORGANIC CHEMISTRY CHAP.

and hydrogen peroxide. These three substances were differentiated
by the following reactions :

1. Caro's acid liberates iodine from potassium iodide instantly.

2. Marshall's acid liberates iodine from iodides only slowly.

3. Hydrogen peroxide at once reduces potassium permanganate,

whilst this is not changed by per sulphuric acids.

In the solution they determined the ratio SO3 : peroxide O, and
found this to be 1 : 1; hence the formula of Caro's acid is SO3 -j- O +
H2O, or H2S05. The free acid was prepared in a pure state by Ahrle
(1909) by the action of sulphur trioxide on anhydrous hydrogen
peroxide : S03 + H202 = H2S05. The reaction with concentrated
sulphuric acid is reversible : H2S04 -f- H202 — H2S05 + H2O.
Caro's acid is crystalline, melts at 45°, and is stable for some days.
D'Ans and Friedrich (1910) prepared both Caro's acid and Marshall's
acid by the action of hydrogen peroxide on chlorosulphonic acid :

HO-SO2-C1 + H202 = HO-S02'0'OH -f HC1

HO S02-C1 + HO-S02-0-OH = HO'S02-0-0-S02'OH + HC1.

H2S2O8 forms crystals stable up to 60°, but in solution slowly passes
into Caro's acid and sulphuric acid : H0O -f H2S208 = H2S04 -j-
H2S05.

The constitution of Caro's acid, or permonosulphuric acid, is seen
by the above reactions to be

/0-OH
SO /
X)H

whilst that of perdisulphuric acid, or the ordinary acid, is :

OS02-OH

•S02-OH

These formulae agree with the constitution adopted for hydrogen
peroxide.

THIOSULPHURTC AND THIONIC ACIDS.

Thiosulphuric acid. — If a solution of sodium sulphite is boiled with
flowers of sulphur, a salt separates on evaporation and cooling
which has the formula Na2S2O3,5K2O. This may be regarded as
sodium sulphate in which an atom of oxygen is replaced by one of
sulphur, and is hence known as sodium thiosulphate, It is commonly
called sodium " hyposulphite," but this name is more appropriately
given to the compound Na2S2O4 (p. 525). On account of the simi-
larity in the chemical properties of oxygen and sulphur, one may
suppose that the sulphur atom in the above reaction : Na2S03 +
S '= Na2S203, enters the molecule of the sulphite in the same position
as the oxygen atom in the reaction Na2S03 + O = Na2S04. The

xxvi THE OXYGEN COMPOUNDS OF SULPHUR 521

formula of sodium sulphite, however, may be either S02/ or

\Na

/ONa

S0<f (p. 496), so that there are two possible formulae for the

XONa

, ,

thiosulphate, viz., S02\ or SO'S

SNa

Spring, by the action of iodine on a mixture of sodium sulphide and
sodium sulphite, obtained sodium thiosulphate. This is a condensa-
tion reaction, and two modes of interaction are possible :
/ONa /ONa

I. SO / = SO,/ + 2NaI

X!Na~NaJNaS XSNa

i I, !
/Na /Net

II. S0 - S02< + 2NaI

It is considered that I. is more probable, since the formula of the
thiosulphate is then more analogous to that of the sulphate. Spring
showed that if the thiosulphate is treated with sodium amalgam and
water, the above condensation reaction is reversed, and sodium
sulphite and sodium sulphide are produced. Further, if sodium
silver thiosulphate, produced when a silver salt is dissolved in a
solution of sodium thiosulphate (p. 828), is boiled with water, a
black precipitate of silver sulphide is produced : SO2(OAg)(SAg) -f-
H20 = S02(OH)2 + Ag2S.

Sodium thiosulphate, Na2S203,5H20, commonly called " hypo,"
is made by boiling the sulphite with sulphur, or by oxidising alkali-
waste, containing calcium disulphide, CaS2, by exposure to air and
then precipitating the calcium with sodium carbonate :

2CaS2 + 302 = 2CaS203, and CaS203 + Na2C03 = CaC03 +
Na2S203.

If sulphur is boiled (or fused) with a caustic alkali, or. milk of lime,
a thiosulphate is produced as well as a sulphide : 6NaOH -f- 4S =
Na2S2O3 + 2Na2S -f 3H2O. Thiosulphates are also formed by
passing sulphur dioxide through solutions of sulphides : the re-
action (which led to the discovery of thiosulphates by Chaussier in
1799), according to Vauquelin probably proceeds in three stages :

1. SO2 + Na2S + H2O = Na2S03 + H2S.

2. SO, -f 2H2S = 2H20 + 3S.

3. Na2S03 + S = Na2S2O3.

522 INORGANIC CHEMISTRY CHAP.

If a solution of sodium thiosulphate is acidified, free thiosulphuric
acid, H2S203 (which is unknown), is probably first formed, but im-
mediately decomposes into sulphurous acid and free sulphur, which
slowly deposits as a white turbidity : H2S2O3 =H2SO3 + S. The
delay in the appearance of the precipitate is due to the formation of a
colloidal solution, and not to the slow decomposition of H2S2O3,
since the sulphur is ultimately precipitated, even if the solution,
after acidification, is at once neutralised with caustic soda, when
any H2S2O3 would be reconverted into Na2S2O3.

Sodium thiosulphate readily dissolves silver chloride, bromide, and
iodide, forming double salts, which have a sweet taste, e.g., NaAgS203.
For this reason the salt is used in photography to remove unaltered
silver halides from the negatives or prints, so as to render these per-
manent to light (" fixing," p. 830).

It is very readily oxidised by chlorine water : Na2S203 + 4HOC1
-f H20 = Na2SO4 + H2S04 + 4HC1, and is used as an antichlor
to remove traces of chlorine from bleached fabrics. With bromine
the reaction is similar, but with iodine an entirely different reaction
occurs.

Tetrathionic acid, H2S406. — If a solution of sodium thiosulphate
is added to a solution of iodine, the colour of the latter is discharged.
This is used in the titration of iodine ; a little starch -paste may be
added when the colour is almost discharged, and the blue colour then
disappears when the last trace of iodine has reacted. The product
of the reaction is not sodium sulphate, but a new salt of the formula
Na2S4O6, sodium tetrathionate ; it was discovered bv Fordos and
Gelis in 1843. The reaction is : 2Na2S203 + Ia = 2NaI + Na2S406.
It is one of condensation :

I"""NaiNaS208 Nal

| + + Na2S406.

I NajNaSa08 Nal

The reaction is quantitative. To obtain the pure salt, a saturated
aqueous solution of sodium thiosulphate is added drop by drop to a
cooled solution of iodine in alcohol. The tetrathionate separates as
it is formed ; it is washed with alcohol, dissolved in water, repre-
cipated with alcohol, and dried over sulphuric acid. In solution
the salt slowly decomposes : Na2S406 = Na2S04 -f SO2 -f 2S ;
the reaction is accelerated by sodium thiosulphate.

If lead acetate is added to a solution of sodium thiosulphate, a
white precipitate of lead thiosulphate is obtained. This, when
suspended in water and treated with iodine, gives a solution of lead
tetrathionate : 2PbS203 + I2 = PbI2 + PbS4O6. When this is
precipitated with sulphuretted hydrogen, a solution of free tetra-
thionic acid, H2S4O6, is obtained. The solution may be concentrated
on a water-bath, and is fairly stable. When concentrated beyond

xxvi THE OXYGEN COMPOUNDS OF SULPHUR 523

a certain point, however, it decomposes : H2S4O6 = H2SO4 -f-
S0a + 2S.

By the action of sodium amalgam and water on sodium tetra-
thionate, the reaction of condensation by which it was formed is
reversed, and sodium thiosulphate is reproduced : Na2S4Ofi + 2Na

Tetrathionates give with sulphides a precipitate of sulphur :
Na2S406 + Na2S = 2Na2S203 + S.

Dithionic acid. — If finely-ground pyrolusite (native crystalline
manganese dioxide) is suspended in water and sulphur dioxide passed
in, manganous sulphate is formed, together with the salt of a new
acid (Gay-Lussac and Welter, 1819). If the liquid, after several
hours' treatment, is filtered and baryta water added, a precipitate of
barium sulphate is formed, and the barium salt of the new acid
remains in solution. On evaporation colourless crystals of barium
dithionate, BaS206,2H2O, separate.

Manganic sulphite is first formed, and then decomposes as
follows :

2Mn02 + 3H2S03 = Mn2(SO3)3 + 3H2O + O.
Mn2(SO3)3 = MnS206 + MnSO3.
" MnS03 + 0 = MnS04.
MnS2O6 + Ba(OH)2 = Mn(OH)2 + BaS206.

By decomposing the barium salt with the calculated amount of
sulphuric acid, a solution of dithionic acid, H2S206, is formed, which
may be concentrated on a water-bath to a certain extent, but then
decomposes : H2S206 = H2S04 -f- SO2. No sulphur is deposited.
The salts decompose on heating in a similar manner : K2S206 =
K2S04 + S02.

On treating sodium dithionate with sodium amalgam, sodium
sulphite is formed, hence the formula of the acid is probably
(SO.-OH), :

S02-ONa Na S

S0-ONa Na '

Trithionic acid. — By the action of heat on a solution of potassium
silver thiosulphate, silver sulphide is precipitated, and the solution
contains the sodium salt of trithionic acid, Na2S3O3 :
/OK

S02<

X:SAe! /SO2-OK

g£3 = Ag2S + S<

Wiiig \Qrk -OTT

SQ / S02 OK

524 INORGANIC CHEMISTRY CHAP.

The same salt is formed by saturating a solution of potassium thiosul-
phate with sulphur dioxide until it is yellow, allowing it to stand till
colourless, and again passing in SO2 : 3SO2 + 2K2S2O3 = 2K2S3OG + S.
The salt crystallises out.

Pentathionic acid. — If sulphuretted hydrogen is passed into a
solution of sulphurous acid, a variety of substances is formed.
Colloidal sulphur is precipitated, and the milky liquid, known as
Wackenroder's solution (1845), contains two new thionic acids,
pentathionic acid, H2S5O6, and hexathionic acid, H2S606. If it is
treated with one-third of an equivalent of caustic potash and allowed
to evaporate spontaneously, a mixture of tetrathionate and penta-
thionate is obtained, which may be separated by recrystallisation
from warm water. The mother liquor on spontaneous evaporation
deposits a crust of a salt richer in sulphur, probably the hexathion-
ate. The crystals of tetrathionate and pentathionate may also be
separated by flotation in a mixture of xylene and bromoform
(CHBr3), of sp. gr. 2*2. K2S4O6 sinks, whilst 'K2S506 rises (cf. p. 9).
The solution \contains, in addition to these two thionic acids,
sulphuric acid and a trace of tri thionic acid.

The reactions leading to formation of Wackenroder's solution have
been represented as follows by Debus (1888) :

I. H2S + 3S02 = H2S406.
II. H2S406 + H2S03 = H2S306 + H2S203.

III. 2H2S306 + 5H2S = H2SO4 +' H2S203 + 5H20 + 8S.

IV. H2S406 + H2S203 = H2S506 + H2S03.

The constitution of the thionic acids has been indicated above. The
formulae at present accepted are those proposed by Mendeleeff and
Blomstrand (1870) :

S02-OH /S02-OH /S02-OH /SO2'OH

I S< S2< S3/

SO2'OH XSO2-OH \SO2-OH \SO2-OH

The alternative formulae proposed by Debus are considered less
probable, although the evidence for each set of formulae is not too
convincing :

SO2-OH OSO2-H S-SOyOH S-SO2-OH S-SO2'OH

SO2-OH S-SO2-OH O-SO2-SH O'SO2'S-H O'SO2-S-H

!! /% -

S S S

Hertlein (1896) found that the polythionates of mercury and silver do
not form complex compounds ; hence it is probable that the metal is
attached to oxygen, as in Blomstrand and Meiicleleeffs formulae,
rather than to sulphur, as in Debus' s formulae, since these metals in

xxvi THE OXYGEN COMPOUNDS OF SULPHUR fuM

combination with sulphur readily form complex compounds (p. 870).
Tetrathionic acid, HO'SO2'S'S>SO2'OH, also corresponds with per-
sulphuric acid, HO'SO2'O*O'SO2'OH, and the tetrathionates form com-
pounds with ammonia, etc., similar to those formed by persulphates ;
e.g., ZnS4Oc,4NH3.

Hyposulphurous acid. — If zinc dust is added to a solution of sulphur
dioxide in absolute alcohol, no hydrogen is evolved, but a salt of the
formula ZnS2O4 crystallises out, which may be dried over concen-
trated sulphuric acid : Zn -f 2S02 = ZnS2O4. This is a salt of
hyposulphurous acid, H2S2O4. The solution of the salt is a powerful
bleaching agent, and also shows very powerful reducing properties.
Thus, it reduces a solution of copper sulphate to a red precipitate of
cuprous hydride, Cu2H2, and precipitates mercury and silver from
their salts. The moist compound rapidly absorbs oxygen from the
air, forming a sulphite.

Sodium hyposulphite (sometimes called hydrosulphite), Na2S204,
is prepared by treating a solution of sodium hydrogen sulphite,
NaHSO3, with zinc dust, in a corked flask. Milk of lime is then
added to precipitate the zinc sodium sulphite which is also formed.

4NaHS03 + Zn = ZnSO3,Na2S03 + Na2S2O4 + 2H20 ;
ZnS03,Na2S03 + 2Ca(OH)2 = Zn(OH)2 + 2CaS03 + 2NaOH.

The double salt is also precipitated if alcohol is added to the
original solution. The filtrate contains the sodium hyposulphite. It
is warmed with a concentrated solution of sodium chloride, and
allowed to cool, when thin vitreous prisms of Na2S204,2H2O separate.
These are washed with aqueous, and then with anhydrous, acetone,
and dried over concentrated sulphuric acid, when anhydrous
Na2S2O4 remains as a white powder, which after drying in a vacuum
at 60° is stable. The hydrate very rapidly absorbs oxygen from the
air : Na2S204 + O2 + H2O = NaHSO3 + NaHS04.

The sodium bisulphite solution may first be saturated with sulphur
dioxide : 2NaHSO3 -j- S02 + Zn = Na2S204 + ZnS03 + H20.
Sodium hyposulphite is also formed when sulphur dioxide, diluted
with nitrogen or under reduced pressure, acts on sodium hydride :
2NaH + 2SO2 = Na2S2O4 -f- H2. With pure gas, explosions occur.

The free acid is formed as a yellow solution by adding oxalic
acid to a solution of the sodium salt. It rapidly oxidises :
2H2S204 + O2 = 2H2O + 4S02.

The composition of the hyposulphites was determined by Bernth-
sen, who showed that, for every two atoms of sulphur in the hyposul-
phite, one atom of oxygen is required to convert it into sulphite
(which may be effected by an ammoniacal solution of copper sul-
phate), and three atoms to convert it into sulphate (which is effected
by a solution of iodine). These results agree with the formula

520 INORGANIC CHEMISTRY CHAP.

S2O3 for the anhydride (H20,S2O3), but not with SO, which was
formerly accepted (H2SO2 = H2O,SO) :

S203 + O = 2S02 S2O3 + 30 = 2S03

2SO + 20 = 2S02 2SO + 4O = 2S03.

Sodium hyposulphite is used to dissolve indigo, a blue colouring matter
C]6H10N2O2, which is insoluble in water ; a colourless solution of indigo -
white, a reduction compound, is formed :

Na2S2O4 + 2H,O = 2NaHSO3 + 2H (nascent)
C16H10N202 + 2H = C16H12N20,.

If a fabric is soaked in the solution, and exposed to air, oxidation
occurs and indigo -blue is deposited in the fibres. A dye of this character
is called a vat- dye, and several other kinds are used besides indigo, so
that sodium hyposulphite is an important salt in colour chemistry.

Sulphoxylic acid, H2SO2, is known only in the form of an organic
compound with formaldehyde : H-COH-NaHSO2,2HaO.

Sulphur sesquioxide. — If flowers of sulphur are added to fused sulphur
trioxide at 10°, blue drops are formed, which solidify to malachite -
blue crystalline crusts. This substance is sulphur sesquioxide, S2O3.
It slowly decomposes into sulphur and sulphur dioxide : 2S2O3 =
3SO2 + S. It dissolves in fuming sulphuric acid to form a blue liquid,
which is also produced by dissolving sulphur in the fuming acid (Bucholz,
1804). Water decomposes the sesquioxide, with separation of sulphur
and formation of sulphuric and thiosulphuric acids. The oxide is not,
therefore, the anhydride of hyposulphurous acid, H2S2O4, as might be
inferred from its formula. The solution of the sesquioxide in fuming
sulphuric acid is used in the manufacture of certain dyes (thiopyrin).

EXERCISES ON CHAPTER XXVI

1. Describe carefully what is observed when : (a) roll sulphur is
heated in a flask to the boiling point ; (b) concentrated sulphuric acid
is heated with copper turnings ; (c) flowers of sulphur are added to
fuming sulphuric acid. What chemical changes are supposed to occur ?

2. How is sulphur dioxide made (a) in the laboratory ; (b) on the
large scale ? What experiments would you perform in order to ascertain
the composition of the gas ?

3. Describe what happens when charcoal is heated with concentrated
sulphuric acid. How would you proceed to separate the gaseous
products of the reaction ?

4. What salts may be produced from sulphur dioxide and a solution
of caustic soda ? What happens when these salts are heated ?

5. What is the constitution of sulphurous acid ? Assuming that
sulphur is sexivalent, what will be its structural formula ? What
salts should be formed (a) on neutralising NaHSO3 with KOH, (b) on
neutralising KHSO8 with NaOH ? If the same salt is produced in
both cases, what conclusion would you draw as to the formula of
sulphurous acid ?

xxvr THE OXYGEN COMPOUNDS OF SULPHUR r.L>7

0. By what method would you measure the rate of production of
sulphuric acid from a solution of sulphurous acid in presence of oxygen ?

7. What is the action of sulphur dioxide on (a) ozone ; (6) iodine
dissolved in water ; (c) iodic acid ; (d) lead dioxide ; (e) sulphuretted
hydrogen ? How would you determine the percentage of sulphur
dioxide in the residual gas from vitriol chambers ?

8. How is sulphur trioxide prepared, and what are its properties ?
What oxides of sulphur exist besides SO2 and SO3, and what is the

1 action of water on them ?

9. Describe the manufacture of fuming sulphuric acid by the contact
process. What compounds of water and sulphur trioxide exist ?

10. Describe the manufacture of sulphuric acid by the chamber
process. What reactions are supposed to occur in the process, and
what experiment may be performed to illustrate these reactions ?

11. What experiments would you perform in order to show that
sulphuric acid is a dibasic acid ? An acid is sometimes defined as
" a hydrogen compound from which the hydrogen can be replaced by
metals." Discuss this.

12. How may sodium sulphite be prepared ? Starting with
sodium sulpliite, how would you prepare : (a) sodium metabisulphite ;
(6) sodium thiosulphate ; (c) sodium hyposulphite ; (d) sodium tetra-
thionate ?

13. What is an acid chloride ? How are the chlorides of sulphurous
and sulphuric acids prepared, and what light do they throw on the
constitutions of the acids ?

14. Describe briefly how you would prepare specimens of : (a) barium
dithionate ; (b) potassium persulphate ; (c) potassium pyro-
sulphate ; (d) sodium tetrathionate.

15. What are sulphonic acids? How, and for what purposes, are
they prepared ?

16. What reactions occur when (a) dilute sulphuric acid, (6) 50 per
cent, sulphuric acid, are electrolysed ? How may the substance
formed in the second case be obtained in a pure state, and how has its
constitutional formula been established ?

17. How are persulphates prepared ? Describe the preparation of
the two persulphuric acids, and describe the method of differentiating
between them.

18. Give reactions in which sulphur dioxide acts (a) as an oxidising
agent ; (6) as a reducing agent ; (c) as an acid anhydride. How is the
bleaching action of sulphur dioxide explained ?

CHAPTER XXVII

SELENIUM AND TELLURIUM

Selenium. — A new element analogous to sulphur was discovered
in 1817 by Berzelius, in the deposit formed in a sulphuric acid
chamber. It was called selenium, from the Greek selene, the moon,
on account of its analogy to tellurium (q.v.).

Selenium occurs in some specimens of native sulphur, particu-
larly Japanese. Metallic selenides also occur, e.g., clausthalile, PbSe,
also Cu2Se and Ag2Se, at Clausthal (Hartz); onofrite, HgSe,4HgS,
in Mexico ; and croolcesite, (Cu,Tl,Ag)2Se, at Skrikerum (Sweden).
It is found in many varieties of pyrites (especially Norwegian),
and thence finds its way into the flue-dust, and the commercial
sulphuric acid. In making salt-cake with this acid, the selenium
passes over as the chloride, SeCl4, into the hydrochloric acid, from
which the element can be precipitated in the form of a red powder by
sulphur dioxide. To prepare selenium from the flue-dust of pyrites
burners, it is digested with a solution of potassium cyanide, when
potassium selenocyanide (cf. KCNS) is formed : KCN + Se -
KCNSe. On addition of hydrochloric acid, selenium is precipitated :
KCNSe + HC1 = KC1 + HCN + Se. It is purified by evaporating
to dryness with nitric acid, when solid selenium dioxide, Se02, is
formed, which can be recrystallised from hot water as selenious acid,
H2Se03. A solution of this is reduced by sulphur dioxide : H2Se03
+ 2S02 + H2O = Se + 2H2SO4. The element is precipitated as
a red powder.

Selenium may also be extracted from the anode-slimes in copper
refining (p. 809), which may contain as much as 96 per cent, of
the element, together with tellurium.

Forms of selenium. — Various modifications of selenium are known :
according to Saunders (1900) these fall into three main groups :—

1. Liquid selenium — an amorphous solid, which may be regarded
as a supercooled liquid of great viscosity. This exists as : (a) Vitreous
selenium, obtained as an opaque lustrous mass, sp. gr. 4-28, almost
black in colour, but giving a red powder, by suddenly cooling melted
selenium. It softens at 50°, and if very rapidly heated to 220° it is

528

CH. xxvii SELENIUM AND TELLURIUM 529

liquid, although viscous. At temperatures above GO-SO0 it changes
fairly quickly, into metallic selenium (q.v. 3). (6) Colloidal selenium,
obtained as a red solution by mixing dilute aqueous solutions of
selenious and sulphurous acids : SeO2 + 2H2SO3 = Se -f 2H2SO4.
The solution slowly deposits (c) amorphous selenium, a red powder,
sp. gr. 4-26, also formed by precipitating a solution of selenium in
potassium cyanide by hydrochloric acid, or by subliming selenium
in a sealed tube. These three varieties dissolve in carbon disulphide.

2. Crystalline selenium, produced from 1 (a) or 1 (c) on standing
in contact with carbon disulphide, by adding benzene to a solution
of selenium in carbon disulphide, or by the spontaneous evaporation
of this solution. Two stable red, monoclinic, crystalline varieties are
known, sp. gr. 4-47 (c/. sulphur). If heated rapidly the crystals fuse
at 200° ; partial conversion into metallic selenium has probably
occurred, and the unstable melting point of the crystals is probably
170-180° (c/. a-sulphur, p. 479).

3. Metallic selenium is formed when any other variety is heated at
200-220° for some time. It is a steel-grey mass, sp. gr. 4-80, giving a
black powder (red if very fine), and is insoluble in carbon disulphide
(about 1 per cent, of soluble selenium is always present).

The boiling point of selenium is 690° ; the vapour is dark red,
and its density diminishes with rise of temperature, becoming
constant (Se2) above 1400°.

t° A(H = 1)
774 101 -2

815 95-4

900-1800 78-0 (Se2 = 78'5)

The molecular weight in solution in phosphorus corresponds
with Se8.

Metallic selenium, which has been heated for some time at 210°,
has the remarkable property of possessing an electrical resistance which
varies on exposure to light, diminishing with the intensity of illumina-
tion (Willoughby Smith, 1873). When the light is cut off, the original
conductivity is recovered after a short time. This effect, which is
utilised in the photophone and other instruments, was attributed by
Siemens (1875) to the existence of two forms of metallic selenium,
one a good conductor of electricity and formed from the other on
exposure to light. These two forms have been isolated. Form A
consists of round granular crystals, stable at 140°, and an insulator in the
dark. Form B, which is produced when Form A is heated to 200°
for some time, or is exposed to light, forms longer crystals, and is a
conductor (Marc, 1903, and Hies, 1908). The action is chiefly produced
by red rays.

M M

530 INORGANIC CHEMISTRY CHAP.

Selenium is used in making red glass, or red enamels and glazes.

Hydrogen selenide, H2Se. — This gas is formed by heating selenium
in a sealed tube with hydrogen : H2 -)- Se ;=± H2Se. Most of the
selenium sublimes in the form of glittering crystals. By heating
iron filings with selenium, iron selenide is formed, which gives H2Se
with acids : FeSe + 2HC1 = FeCl2 -f H2Se. Hydrogen selenide
is a colourless inflammable gas, with a very offensive smell, and a
strong action on the mucous membranes. It is soluble in water,
giving a feebly acid solution which precipitates selenides of many
metals, and oxidises on exposure to air, selenium being precipitated.
The density of the gas is 40-7, and it leaves its own volume of hydro-
gen when decomposed by heated tin ; hence its formula is H2Se.
It liquefies at — 42°, and solidifies at — 64°.

No perselenides of hydrogen are known.

Halogen compounds of selenium. — Selenium forms two fluorides,
SeF4 and SeF6, and two chlorides, Se2Cl2 and SeCl4. The dichloride,
Se2Cl2, is formed as a brown liquid by passing chlorine over fused
selenium. It is slowly decomposed by water : 2Se0Cl2 -j- 3H2O -
HoSeO3 + 3Se 4- 4HC1. On heating it decomposes : 2Se2Cl2 -
3Se + SeCl4. The tetrachloride is therefore more stable than Se2Cl2
(cf. S2C12 and SC14) ; it is produced as a white solid by treating the
dichloride with chlorine, or by heating Se02 with PC15 : 3SeO2 -f
3PC15 = 3SeCl4 + P2O5 4- POC13. It sublimes without melting,
and its vapour is dissociated : 2SeCl 4 ^=± Se2 -}- 4C12 (Evans and
Ramsay), or 2SeCl4 ^± Se2012 + 3C12 (Chabrier). It is decomposed
by water : SeCl4 + 3H2O = 4HC1 + H2Se03. By the action of
SeCl4 on Se02, a yellow liquid oxychloride, SeOCl2, is formed.
Se2Br2 and SeBr4 are known, but the iodides appear to be mixtures.

Oxides and oxy -acids of selenium. — Selenium bums in oxygen
with a blue flame, producing a crystalline dioxide, Se02. Indications
of the existence of a second solid oxide, Se304, have been obtained.
A trace of a gaseous oxide ( ? Se03) seems to be produced during the
combustion of selenium ; it possesses a strong odour of horse-radish.
A similar smell, due to carbon diselenide, CSe2, is emitted when
selenium is heated on charcoal before the blowpipe.

If SeO2 is dissolved in hot water, or selenium boiled with nitric
acid, colourless prismatic crystals of selenious acid, H2Se03, separate
on cooling. It is a dibasic acid, forming acid and normal
salts, e.g., KHSe03, K2Se03. Superacid salts are also formed :
KHS03,H2Se03. It is readily reduced (e.g., by organic matter in
dust) with deposition of selenium. Potassium permanganate oxidises
selenious to selenic acid.

Selenium trioxide is not known, but selenic acid, H2Se04, is
produced by the action of chlorine on selenium, or selenious acid,
suspended in water : Se + 4H20 + 3C12 == H2Se64 + 6HC1 ; by the
action of bromine on silver selenite in water : Ag2Se03 + H20 +

xxvii SELENIUM AND TELLURIUM 531

Br2 = 2AgBr -f H2Se04 ; or by the electrolytic oxidation of a solu-
tion of selenious acid in nitric acid. The solution may be evaporated
until, at 265°, it contains 95 per cent, of H2SeO4, which decomposes
on further heating. If this liquid is placed over sulphuric acid in
an evacuated desiccator until it contains 974 per cent, of H2SeO4
(sp. gr. 2-627), and is then strongly cooled, crystals of pure selenic
acid (m.-pt. 58°) separate. The acid is very hygroscopic and
evolves heat when mixed with water ; the strong solution chars
organic matter. The potassium salt is formed on fusing potassium
selenite with nitre (Mitscherlich, 1827), the sodium salt from
selenium and sodium peroxide. Selenic acid is also formed by treat-
ing Se02 with acidified permanganate solution.

The heated acid dissolves copper and gold, producing SeO2, and
selenates. The dilute acid dissolves zinc, iron, etc., liberating
hydrogen and forming selenates. Barium selenate is sparingly
soluble in water.

Selenic acid is not reduced by sulphur dioxide, or by sulphuretted
hydrogen, but it is decomposed, with formation of selenious acid, by
boiling with hydrochloric acid, even in dilute solution : H2SeO4 -j-
2HC1 = H2Se03 + C12 + H2O. The solution then deposits selenium
when treated with sulphur dioxide.

Selenium dissolves in fused sulphur trioxide, or oleum, the com-
pound SSe03 (selenosulphur trioxide) being formed in green crystals.
(Sulphur gives blue S203 ; tellurium bright red STeO3.)

Selenium dissolves in potassium sulphite solution, giving a pink,

/OK
unstable solution of the selenosulphate, SO (cf. K2S2O3).

Organic compounds of selenium (e.g., selenium indigo) are used in
destroying cancer cells.

A very delicate test for selenium is the formation of a blue colour
with /3-imino-a-cyano-hydrindene dissolved in concentrated sulphuric
acid r commercial sulphuric acid usually contains sufficient selenium
dioxide to give this reaction.

Tellurium. — Tellurium occurs in small quantities in the free state,
and was called by the early mineralogists aurum paradoxum,
or metallum problematum, on account of its lustre. Miiller von
Reichenstein (1782) concluded that it was a peculiar metal ; it
was more carefully examined by Klaproth (1798), who called it
tellurium, Berzelius (1832) pointed out its analogies with sulphur
and selenium, placing the three elements in the same group. It is
now usually regarded as a non-metal.

Tellurium occurs only in relatively small quantities ; native
tellurium is found in Central Europe, America, and Bolivia, but the
element more usually occurs in combination with metals as tellurides :

M M 2

532 INORGANIC CHEMISTRY

graphic tellurium (or xi/lrtniitc}, (Ag,Au)Tt\, : hln-ck
(Au,Pb)2(Te,S,Sb)3 ; hestite, Ag2Tc ; tetradymite, 15i2rJV3, etc. It
is present, together with selenium, in Japanese sulphur.

Tellurium is usually extracted from the residues from bismuth
ores. These are dissolved in hydrochloric acid, and sodium sulphide
is added. Tellurium is precipitated. It is purified by boiling with
sodium sulphide solution and powdered sulphur, then adding sodium
sulphite : tellurium separates as a greyish-black precipitate, which
becomes silver- white on fusion. It crystallises in rhombohedra, is
brittle and easily powdered, and has a fairly high sp. gr. of 6-27.
It conducts electricity like a metal. An amorphous variety (sp. gr.
6-015) is precipitated by sulphur dioxide from tellurous or telluric
acid.

Tellurium melts at 452°, and boils at 1400° ; in a nearly perfect
vacuum it boils at 478°, forming a golden-yellow vapour. The
vapour density at 1400° is slightly higher than that corresponding
with Te2. It burns with a blue flame when heated in air, forming
white vapours of tellurium dioxide, Te02, which is also formed on
treating tellurium with nitric acid, or by heating the basic nitrate.
Te02 occurs native as tellurite ; it is only sparingly soluble in water,
the solution giving no acid reaction with litmus. Te02 is in fact also
a weak base, forming salts derived from Te(OH)4, e.g., the basic
nitrate, 2Te02,HNO3.

Hydrogen telluride, H2Te. — This combustible gas was prepared
in an impure state by Davy in 1810 by treating zinc telluride with
acids ; pure H2Te is obtained from aluminium telluride and dilute
hydrochloric acid or by the electrolysis of 50 per cent, sulphuric or
phosphoric acid with a tellurium cathode, at once drying the gas, and
cooling to — 20°. It is then obtained as a liquid, b.-pt. 0°, m.-pt.
- 48°. It is fairly stable in the dark, but on exposure to light,
especially in presence of moisture, it decomposes : H2Te = H2 -f Te.
The vapour density is 65-1, and the volume is unchanged on heating
with zinc, hence the formula is H2Te. By allowing an aqueous
solution of H2Te to oxidise in the air, a claret-red solution of colloidal
tellurium is formed.

Tellurium, when fused with potassium cyanide, does not form any
compound analogous to KCNS or KCNSe, but only K2Te.

Halogen compounds of tellurium. -Tellurium" dichloride, TeCl2,
is formed as an indistinctly crystalline black mass by passing chlorine
over melted tellurium. It gives a deep red vapour, which becomes
yellow in air, TeOCl2 and TeCl4 being formed. It is decomposed by
water : 2TeCl2 + 3H2O = Te + H2Te03 + 4HC1. With excess of
chlorine the stable white crystalline tetrachloride, TeCl4 (m.-pt.
224°) is formed. This is very hygroscopic, and is hydrolysed by
water, producing tellurous acid, H2TeO3 : TeCl4 + 3H20 ^H2TeO3
+ 4HC1. The vapour is stable up to 530°. The iodide" is formed in

xxvii SELENIUM AND TELLURITIM 533

iron-grey crystals by the reaction : H2Te03 + 4HI = TeI4 -f 3H20.
TeF4, TeF6, TeBr2, TeBr4, are known.

Telluric acid. — Tellurium trioxide, TeO3, is obtained by heating tel-
luric acid, H2Te04. It is an orange-yellow powder, which decomposes
when strongly heated : 2TeO3 = 2TeO2 -f- O2. It does not recombine
with water. Telluric acid is a very weak acid, formed by dissolving
tellurium in a mixture of nitric and chromic acids, washing the crystals
with nitric acid, and recrystallising from water. It forms white
crystals of the composition H2TeO4,2H2O. These, unlike true crystal-
line hydrates (e.g., CuSO4,5H2O), are not permeable to water-
vapour in thin plates, hence they appear to have the formula
Te(OH)6. The molecular weight in solution also corresponds with
this formula. The acid is dimorphous ; below 10° it forms
H2Te04,6H60. It is difficultly soluble in cold water, but readily dis-
solves in hot water. The methyl ester, Te(OCH3)6, is also known. When
heated to 160°, H6Te06 forms allotelluric acid, H2Te04. Metallic
tellurates are formed by fusing tellurites, e.g., K2TeO3, with nitre,
or passing chlorine through their alkaline aqueous solutions :
K2Te03 + 2KOH + C12 = K2Te04 + 2KC1 + H20. They are not
isomorphous with the sulphates, although the acid selenates and
tellurates of rubidium are isomorphous. Some tellurates exist in
two forms, a colourless salt soluble in water and acids, and a yellow
insoluble form. Normal and acid salts and complex superacid
salts (e.g., K2Te04,Te30,4H2O ; K2Te04,3Te03,4H20) are known.
Tellurates are reduced to tellurites on boiling with hydrochloric
acid : K2Te04 + 2HC1 = K2Te03 -f H20 + C12. Barium tellurate
is sparingly soluble in water.

If the red compound STeO3 (p. 531) is heated in vacuo to 230°, SO2
is evolved and a brownish-black mass of the monoxide, TeO, is left.
This dissolves in concentrated sulphuric acid, forming a crystalline
mass of tellurous sulphate, Te(SO4)2.

The atomic weight of tellurium. — The anomalous positions ol
iodine and tellurium in the periodic system led to the suspicion that
tellurium might contain an unknown element of higher atomic
weight. Brauner (1883) attempted to separate this, and believed
that by distilling tellurium in hydrogen its atomic weight was
reduced from 127-6 to 125-57. In this case it would correspond with
its position in the periodic table.

H. B. Baker and A. H. Bennett (1907) attempted to separate the
supposed constituents : (1) by fractional crystallisation of telluric
acid ; (2) by boiling barium tellurate with water (the solubility in-
creases in the series BaSO4 -> BaSe04 -> BaTe04) ; (3) by fractional
distillation of Te, Te(C2H5)2, TeCl4, and TeO2 ; (4) by fractional
electrolysis of tellurium compounds ; (5) by fractional precipitation
of TeCl4 with water. The results were all negative. By heating

534 INORGANIC CHEMISTRY CH. xxvn

Te02 with sulphur in a small tube (Fig. 281) the reaction Te02 -f
S = Te + SO2 occurred, the excess of sulphur being kept back
with silver foil. By this method, and the
synthesis of TeBr4, the value Te = = 126-5
(H = 1) was obtained, which is higher than the
atomic weight of iodine 1= 125-91. Flint
(1909) claimed to have succeeded in separating
tellurium by method (5), but this has not Ixvn
substantiated by Harcourt and Baker. The
value at present accepted is Te = 126-5.

EXERCISES ON CHAPTER XXVII

1. From what sources is selenium obtained ?
Describe the properties of the element. For

Weight of Tellurium0 what purpose is it used ?

2. Describe the preparation and properties <>i

the important halogen compounds of selenium and tellurium. Contrast
their properties with those of the corresponding compounds of sulphur.

3. How are the oxides and oxy- acids of selenium and tellurium
prepared ? How do they resemble, or differ from, those of sulphur ?

4. Discuss the question of the relative atomic weights of iodine and
tellurium from the point of view of the Periodic Law. How has the
atomic weight of tellurium been determined ?

5. Describe the preparation and properties of the hydrogen compounds
of selenium and tellurium. How have their formulae been established ?

CHAPTER XXVIIT

NITROGEN AND ITS COMPOUNDS WITH HYDROGEN

Nitrogen. — Scheele (1772) first clearly recognised that air is a
mixture of two gases, one of which (fire air] supports combustion and
respiration, whilst the other (foul air) does not. Lavoisier's
(1775-6) experiment (p. 47) furnished a decisive proof of this result,
although both gases had been separately prepared by Scheele. The
latter also showed that, when they were mixed in proper propor-
tions, common air was formed. Lavoisier gave to Scheele's foul air
the name azote (Greek a, no ; zoe, life), which is still used in France ;
the name nitrogen (Greek nitron, nitre), suggested by Chaptal, is
now used elsewhere for the gas.

In 1772 Daniel Rutherford allowed mice to breathe in air under a
bell -jar, and removed the fixed air by washing the residual gas with
potash. A gas remained, which he called mephitic air, since it did
not support combustion or respiration ; unlike fixed air, it was not
absorbed by alkali or lime-water. Priestley (1772) burnt charcoal
in a confined volume of air, and absorbed the fixed air with alkali,
also obtaining mephitic air, which he called phlogisticated air. Both
these experimenters considered that the gas was common air saturated
with phlogiston, or phlogistic material, emitted by the animal or com-
bustible body.

Atmospheric nitrogen was considered to be a pure substance
until 1894, when Rayleigh and Ramsay found that it contained
a little more than 1 per cent, by weight of an inert gas which,
unlike nitrogen, did not combine with heated magnesium. The
existence of this gas had been suspected by Cavendish in 1785.
The inert gas was called argon (Greek argon, sluggish) ; later experi-
ments (p. 603) showed that it contained, besides argon, traces of
other inactive gases : helium, neon, krypton, and xenon.

The composition of air, freed from moisture and carbon dioxide,
is roughly 4 volumes of nitrogen to 1 volume of oxygen ; the exact
figures (Leduc, 1896) are :

By weight. By volume.

Nitrogen 75-5 78-06

Oxygen 23-2 21-00

Argon, etc. .. .. 1-3 0-94

535

536

INORGANIC CHEMISTRY

Cavendish found that the composition of the air is sensibly
constant : 20-833 vols. of oxygen and 79-167 vols. of nitrogen
(including argon).

The composition of air, however, is slightly variable : the per-
centage of oxygen varies from 20-26 to 21-00, according to the
locality, etc., and it is variable at different times at the same place.
Air is thus entirely unsuitable as a standard of relative density.
Traces of free nitrogen are found in volcanic gases, and in the
gases evolved from coal.

Combined nitrogen is widely distributed, and is a constituent

of some of the most impor-
tant compounds. In com-
bination with hydrogen it
forms the base ammonia,
NH3, occurring in the free
state, and as salts in air,
in water, and in volcanic
districts. In combination
with oxygen, nitrogen forms
nitrous, HN02, and nitric,

acids, salts of which
are fairly abundant. Exten-
sive deposits of sodium
nitrate occur in Chile.
Animal and vegetable
organisms contain complex
organic substances called
proteins, • containing an
average of 16 per cent, of
nitrogen. Combined nitro-
gen is a constituent of
explosives such as gunpowder,
nitroglycerine, gun-cotton,
T.N.T., and picric acid ; of
drugs such as antipyrine. and
alkaloids such as quinine and morphine ; and of colouring matters such
as indigo, and aniline dyes. Although free nitrogen is one of the most
inert elements, its compounds exhibit a most wonderful diversity of
properties, and enter readily into chemical reactions. The chemistry
of nitrogen is therefore a subject of great interest and importance.
Preparation of nitrogen from air. — Nitrogen may be prepared :
(a) from air, by removal of oxygen, (6) from nitrogen compounds.
That obtained from air, called atmospheric nitrogen, is not quite
pure, since it contains about 1 per cent, of inactive gases, which
give it a slightly higher density than pure or chemical nitrogen,
prepared from compounds.

FIG. 282. — Preparation of Nitrogen from Air by
passing it over Copper Turnings moistened with
Ammonia.

NITROGEN AND ITS COMPOUNDS

537

XXVIII

Atmospheric nitrogen is produced by the action of phosphorus,
moist iron filings, liver of sulphur, etc., on air at the ordinary
temperature (p. 40). Phosphorus in the form of wire removes
atmospheric oxygen completely. The
oxygen is also removed by an alkaline
solution of pyrogallol, by an acid solution
of chromous chloride (p. 166). or by a
solution of cuprous chloride in hydro-
chloric acid or ammonia : 4CuCl -f-4HCl-f-
O2 = 4CuCl2 -f 2H2O. Metallic copper in
contact with hydrochloric acid or ammonia
also removes oxygen from air.

EXPT. 194. — Pack a drying tower (Fig.
282) with clean copper turnings. Fit the
upper outlet with a dropping funnel and
a tube leading to a wash-bottle and
pneumatic trough. Allow concentrated
ammonia to drop over the copper turnings,
and pass a slow stream of air upwards
through the tower. The nitrogen passing
on is washed with dilute sulphuric acid.
A deep blue solution of cupric oxide in

ammonia is formed, and may be run off from time to time by a stopcock
at the base of the tower. The gas contains a trace of oxygen, which
may be removed by a solution of chromous chloride. (Berthelot.)

Oxygen is also removed from air by burning phosphorus, but not
completely.

FIG. 283. — Burning Phosphorus
in Air.

FIG. 284.— Gravimetric Composition of Air (Dumas and Boussingault).

538 INORGANIC CHEMISTRY CHAP.

EXPT. 195. — Float a porcelain capsule containing a piece of phos-
phorus on water, kindle the phosphorus with a hot wire, and cover
with a bell-jar divided, from the water-level, into five equal volumes
by strips of label (Fig. 283). At once insert the stopper. When the
phosphorus ceases to burn, the fumes of phosphorus pentoxide, P2O5,
dissolve in the water. Allow the apparatus to cool, and equalise the
water-levels. The residual gas occupies four volumes, and will be found
to extinguish a lighted taper.

Oxygen is removed from air by passing the latter, dried, and
freed from carbon dioxide, by solid caustic potash, over copper
turnings heated to redness in a hard glass tube. From the increase
in weight, due to the formation of oxide of copper, the amount
of oxygen in a given volume of air, passed over from a gas-holder,
may be determined. The nitrogen may be collected in an evacuated
globe and weighed, and thus a gravimetric analysis of air carried
out. The apparatus is shown in Fig. 284. In this way Dumas
and Boussingault (1841) found that 100 parts of air contained
23-00 parts of oxygen and 77-00 parts of nitrogen by weight.

If air is bubbled through a warm concentrated solution of ammonia
and the gas passed over a mixture of copper turnings and copper
oxide heated to redness in a hard glass tube, the hydrogen of the
ammonia is burnt by the oxygen of the air : 4NH3 + 302 =
2N2 + 6H20. The gas so prepared (Vernon Harcourt) is a mixture
of atmospheric and chemical nitrogen : its density is intermediate
between the densities of these two gases.

Nitrogen is manufactured either by passing air over red-hot
copper, or by the fractionation of liquid air. The latter method,
described on pp. 175-7, is now mostly used.

Preparation of nitrogen from its compounds. — Nitrogen may be
obtained by the complete oxidation of ammonia : 4NH3 -f- 302 =
2N2 -\- 6H2O. The oxidation may be effected by a hypochlorite or
hypobromite (p. 401) : 3NaOCl -f 2NH3 - 3NaCl + 3H2O + N2.

EXPT. 196. — To 100 c.c. of concentrated ammonia in a flask add
gradually a thin paste of 40 gm. of bleaching powder, with a little milk
of lime, through a thistle funnel. Nitrogen is evolved, with frothing,
on warming : 3Ca(OCl)2 + 4NH3 - 3CaCl2 + 6H2O + 2N2.

EXPT. 197. — Add 6 c.c. of bromine to a solution of 10 gm. of caustic
soda in 100 c.c. of water, cooling by running water. The solution of
sodium hypobromite and bromide is placed in a flask and ammonia
solution dropped in ; or the hypobromite is dropped on solid ammonium
chloride : 3NaOBr + 2NH3 = 3NaBr + 3H2O + IjT2. Nitrogen is also
evolved by the action of alkaline hypobromite solution on urea :
CON2H4 + 3NaOBr = CO2 + N2 + 2H2O + 3NaBr. The gas con-
tains a trace of nitrous oxide, N2O, which is removed by passing over
red-hot copper.

xxvni NITROGEN AND ITS COMPOUNDS 539

A very convenient method for the preparation of nitrogen is the
decomposition of a solution of ammonium nitrite, by heat :
NH4NO2 = N2 + 2H2O. This takes place only very slowly in a
faintly alkaline solution, but readily if the solution is faintly
acid, so that the reaction appears to be due to the oxidation effected
by free nitrous acid : HNO2 -f- NH3 = N2 + 2H20.

EXPT. 198. — Dissolve 30 gm. of sodium nitrite in the smallest possible
amount of cold water, and add a cold saturated solution of 22 gm. of
ammonium chloride: NaNO2 + NH4Cl^±NaCl + NH4NO2. Filter
from the sodium chloride. Make 5 c.c. of the solution mixed with 20 c.c.
of water faintly alkaline with a drop of dilute ammonia, and another
5 c.c. -f 20 c.c. of water faintly acid with a drop of dilute sulphuric
acid. Heat both solutions and observe the results. Heat the main
quantity of the ammonium nitrite solution in a flask, and collect the
gas over water. The gas contains a little nitric oxide, NO : it is
purified by passing through potassium dichromate solution acidified
with dilute sulphuric acid, and then over heated copper. The sodium
nitrite and ammonium chloride solution may be mixed and heated
directly. Pure nitrogen is also produced by passing a mixture of nitric
oxide and ammonia gas through a red-hot tube :
6NO + 4NH3 = 5N2 + 6H2O.

EXPT. 199. — If red crystals of ammonium dichromate are gently
heated they undergo rapid decomposition, with evolution of nearly
pure nitrogen and steam, leaving a voluminous green residue of
chromium sesquioxide : (NH4)2O2O7 = Cr2O3 + 4H2O + N2.

Nitrogen is produced by the action of chlorine (or bromine) on a
solution of ammonia : the reactions are usually given as :

2NH3 + 3C12 - 6HC1 + N2

6HC1 + 6NH3 = 6NH4C1.

The very explosive nitrogen trichloride is formed as an intermediate
product : 2NH3 + 6HOC1 ^± 2NC13 + 6H2O ; NC13 + 4NH3 =
N2 + 3NH4C1. This substance is formed as a violently explosive oily
liquid by the prolonged action of chlorine on ammonia.

EXPT. 200. — Pass a slow stream of chlorine through a wide tube
into a concentrated solution of ammonia (sp. gr. 0-88) in a Woulfe's
bottle (Fig. 285). As each bubble of gas passes through the liquid,
there is a feeble yellow flash of light, followed by the production of
dense white fumes of ammonium chloride and a brisk evolution of
nitrogen. The gas so prepared contains a little oxygen : it may be
passed through a second Woulfe's bottle filled with broken glass
moistened with water, to filter oft NH4C1 fumes. The experiment
should be interrupted after a very short time, as explosive NC13 is formed
when ammonia is no longer in excess.

540 INORGANIC CHEMISTRY^ < H.M-.

Properties of nitrogen. — Nitrogen is a colourless, odourless, taste-
less gas ; it does not support combustion, or respiration, although
it is not poisonous ; it does not turn lime-water milky. It is
sparingly soluble in water, and has no action on litmus. Nitrogen
can be liquefied by cooling and pressure : its critical tempera-
ture is — 146° ; the critical pressure is 33 atm. The liquid
is colourless, b.-pt. -- 195-7°, sp. gr. at b.-pt. 0-8103. On rapid

FIG. 285. — Decomposition of Ammonia by Chlorine.

evaporation under reduced pressure it forms an ice-like solid,
m.-pt. - 210-5°/86 mm. The normal density of the pure gas
is 1-25107 gm./lit., whilst that of atmospheric nitrogen is 1-25718
gm./lit., i.e., 048 per cent, heavier.

Nitrogen is an inert element, but it combines directly with
oxygen, hydrogen, boron, silicon, tungsten, titanium, manganese,
vanadium, calcium, barium, magnesium, and lithium. In presence
of alkalies, or baryta, it also combines at high temperatures with
carbon to form cyanides ; e.g., NaCN. Compounds of elements
with nitrogen are called nitrides, e.g., Li3N, Ca3N2, Mg3N2, BN.
In these compounds nitrogen is tervalent :

/Li /N = Ca

N— Li, B:N, Ca<
\Li ' \N = Ca

xxvin NITROGEN AND ITS COMPOUNDS- 541

Oxygen and hydrogen combine with nitrogen on sparking ; the
remaining nitrides are formed by passing nitrogen over the element
heated to dull redness. By the action of water they give ammonia •
Ca3N2 + 6H20 == 2NH3 + 3Ca(OH)2. When nitrogen is passed
over strongly-heated calcium carbide it is rapidly absorbed,
with formation of a mixture of calcium cyanamide and graphite'
CaC2 + N2 == CaCN2 + C.

EXPT. 201.— Burn a piece of magnesium ribbon in air. Heat the
white product, containing MgO and Mg3N2, with water in a test-tube,
and hold a piece of moist red litmus paper in the tube. It is turned
blue by the ammonia evolved.

EXPT. 202. — Heat some magnesium powder in nitrogen in the short
limb of a bent hard glass tube over mercury. The mercury slowly
rises, owing to absorption of nitrogen.

Active nitrogen. — Just as ozone is produced from oxygen by the
action of an electric discharge, an active form of nitrogen is obtained
by subjecting a current of nitrogen, drawn through a tube at low
pressure, to a high tension discharge. The gas travelling beyond the
portion of the tube in which the discharge occurs glows with a greenish-
yellow light. White phosphorus is converted into red phosphorus,
and sodium and mercury form compounds at 150° when exposed to the
gas. Nitric oxide forms nitrogen dioxide. Strutt (1911), to whom
these discoveries are due, regards the active nitrogen as monatomic,
since the gas is not condensed in liquid air. Recent experiments indi-
cate that a trace of oxygen is necessary in the production of active
nitrogen, although an excess destroys it. It is without action on
hydrogen.

Compounds of nitrogen and hydrogen. — Nitrogen forms three
well-defined compounds with hydrogen :

Ammonia, NH3 ;

Hydrazine, N2H4 ;

Hydrazoic acid, N3H.

The compounds N2H2(di-imide), andN4H4 (buzylene), are known only
in organic derivatives. Compounds N4H5 and N-H5 exist as salts
of hydrazoic acid (p. 559).

Ammonia and hydrazine are basic substances, combining with
acids to form ammonium and hydrazine salts ; e.g.,' NH3-HC1
or NH4C1; N,H4-HC1 or N2H5C1, and N2H4-2HC1 or N2H6C12.
Hydrazoic acid is an acid, dissolving metals with evolution of
hydrogen, and forming salts which are ionised in solution, e.g.,
NaN3 ^ Na' + N3'. The ion N3' is univalent.

542 INORGANIC CHEMISTRY CHAP.

If the hydrogen atoms of ammonia are replaced by hydroxyl
groups, the following compounds are obtained :

NH2OH, hydroxylamine, a base, forming salts, e.g., NH2-OH,HC1 ;
NH(OH)2, dihydroxylamine, unknown in the free state ;
N(OH)3, orthonitrous acid, the hypothetical ortho-acid corre-
sponding with ordinary nitrous acid, HN02.

Ammonia, NH3.— Traces of ammonia occur in the atmosphere :
bottles containing hydrochloric acid become coated after a time
with ammonium chloride. Ammonium chloride, NH4C1, and sulphate,
(NH4)2SO4, occur in volcanic districts ; ammonia also accompanies
boric acid in the fumaroles of Tuscany, and may have been
formed by the decomposition of boron nitride : BN -{- 3H2O =
H3B03 -f- NH3. Small quantities of ammonium salts occur in
plants and animals (e.g., in blood, and in urine as microcosmic salt,
NaHNH4P04), in the soil, and in natural waters (as nitrite and
nitrate).

Ammonia is obtained as a by-product in the destructive distilla-
tion of organic matter containing nitrogen (coal, horn, bones, etc.).
On the small scale, the yield of ammonia is greater if the materials
are mixed with soda-lime, prepared by slaking quicklime with
caustic soda solution, and heating till dry.

EXPT. 203. — Heat a few pieces of feather, or a little glue, with soda-
lime in a test-tube, and test the vapours : (a) with moist red litmus
paper, which is turned blue ; (6) with a glass rod dipped in concentrated
hydrochloric acid, which evolves white fumes of ammonium chloride
NH4C1 ; (c) with paper dipped in mercurous nitrate solution, which is
turned black. Repeat the experiment with filter-paper (free from
nitrogen) without soda-lime, but test with blue litmus paper. In this
case acetic acid, C2H4O2, is formed.

Ammonium chloride, NH4C1, is described by the Latin Geber, and
was called sal armoniacum. It appears to have been derived from
the volcanoes of Central Asia. Later, it was brought from Egypt,
and seems to have been prepared from the soot formed on burning
camels' dung. Its name was changed to sal ammoniacum, previously
given to common salt found in the Libyan Desert near the ruins of
the temple of Jupiter Ammon (Greek ammos — sand). This name was
subsequently abbreviated to sal ammoniac.

A solution of ammonium carbonate, (NH4)2CO3, was also obtained by
distilling putrefied urine : CON2H4 (urea) + 2H2O = (NH4)2CO3 ; or
by the dry distillation of bones, hoofs, horns, etc. ; it was known as
spirit of hartshorn, sal volatile, or the volatile alkali. By distilling this
with quicklime, a solution of caustic volatile alkali, NH4OH, was
obtained, described in Kunckel's posthumous works (1716). Gaseous

XXVITI NITROGEN AND ITS COMPOUNDS , 54:i

ammonia was first obtained by Priestley in 1774, by collecting over
mercury ; he called it alkaline air, and found that when electric sparks
were passed through it double the volume of a combustible gas was
formed : 2NH3 = N2 + 3H2. Berthollet (1785) showed that nitrogen
and hydrogen were formed in this decomposition; the result was
confirmed, and the formula NH3 established, by Austin (1788), Davy
(1800), and Henry (1809).

Ammonia is formed from its elements when these are sparked
together : N2 -f 3H2 ^ 2NH3. This appears to have been dis-
covered by Regnault (1840) ; it was confirmed by Deville (1874),
who pointed out that sparks will bring about both the formation
and the decomposition of ammonia. The reaction is reversible,
and a state of equilibrium is set up in which 6 per cent, of NH3
exists with 94 per cent, of the uncombined gases.. If the mixture
N2 + 3H2, and pure ammonia, respectively, are exposed to pro-
longed sparking, contraction ensues in the first case and expan-
sion in the second, until the volumes and compositions are the
same :

2NH3 — N2 + 3H2.

6 per cent. 94 per cent.

EXPT. 204. — Spark a mixture of nitrogen and hydrogen over mercury
in a eudiometer containing a little concentrated sulphuric acid. Observe
the gradual contraction, owing to formation of ammonia, which is
withdrawn by the sulphuric acid.

Synthetic ammonia. — The direct combination of nitrogen and
hydrogen is utilised in the Haber process (1905) for the synthetic
production of ammonia. Since a diminution of volume occurs
in the reaction : 2N2 -f 3H2 = 2NH3, the amount of ammonia
formed in equilibrium will increase with the pressure.

Since heat is evolved in the reaction, the amount of ammonia in
the equilibrium state will diminish with rise of temperature. At
very high temperatures (above 1000°) heat seems to be absorbed
in the reaction, and the amount of ammonia then increases with
the temperature. This explains its formation in the electric
spark.

In order to obtain appreciable amounts of ammonia, the mixture
of nitrogen and hydrogen, which must be very pure, is circulated
by pumps, under 100-200 atm. pressure, or even 1000 atm. in
Claude's process, over a catalyst, which may be a mixture of finely-
divided iron and molybdenum, and the ammonia formed in each
circulation is removed by cooling and liquefaction, or by absorp-
tion in water. The argon present in the atmospheric nitrogen,
which accumulates, is blown off from time to time. The per-

544 s INORGANIC CHEMISTRY CHAP.

centages of ammonia, by volume, present in equilibrium under
various conditions are given in the table below :

Pressure in Temperature °.

atm. 550 650 750 850 950

1 0-077 0-032 0-016 0-009 0-005

100 6-7 3-02 1-54 0-874 0-542

200 11-9 5-71 2-99 1-68 1-07

In 1910 the Haber process was adopted by the Badische Co. in Ger-
many ; in 1916 the production of ammonium sulphate, of the highest
degree of purity, was 500,000 tons annually, at a cost of £6 per ton,
as compared with about £10 per ton by other methods.

The cyanamide process. — Another process which is largely used
for the fixation of atmospheric nitrogen is the cyanamide process of
Frank and Caro (1895).

Nitrogen is passed over crushed calcium carbide with some
calcium chloride or fluoride, heated to 1100°, either by carbon rods
heated electrically inside drums of carbide, or by dropping the
carbide continuously through electric arcs. Calcium cyanamide
mixed with graphite is formed as a dark grey mass : CaC2 +
N2 = CaCN2 + C. This substance is a derivative of cyanamide,
the amide of hydrocyanic acid, i.e., hydrocyanic acid in which
an atom of hydrogen is replaced by the amino-group :

HCN H> NH2-CN -> NCa-CN or Ca:N-N;C.

Hydrocyanic Cyanamide Calcium

acid cyanamide

The " cyanamide " is agitated with cold water to remove un-
changed carbide, and then stirred with water and a little sodium
carbonate in large iron autoclaves, i.e., pressure digesters, into which
steam is blown until the pressure rises to 3-4 atm. The pressure
then rises automatically to 12-14 atm., owing to production of
ammonia, which is blown off, with some steam, through condensers,
the solution formed being treated in a still with steam to drive out
the gas : CaCN2 + 3H20 = CaC03 + 2NH3. The sludge of calcium
carbonate, lime, and graphite (from the cyanamide) is thrown away.

The Bucher process (1917) consists in passing nitrogen over an inti-
mate mixture of sodium carbonate, charcoal or coke, and iron filings.
Sodium cyanide is formed, the iron acting as a catalyst :
Na2CO3 + 4C + N2 = 2NaCN + SCO.

Steam is then blown over the mass, when sodium formate and ammonia
are produced : NaCN + 2H2O = H-COONa -f- NH3. This process,
owing to technical difficulties, has not been a success.

In the Serpek process, formerly worked in France, aluminium
nitride, A1N, was formed by passing nitrogen over a mixture of coke
and bauxite (native aluminium oxide) in a revolving electric furnace

XXVIII

NITROGEN AND ITS COMPOUNDS

545

at 1800° : A12O3 + 3C + N2 = 2A1N + SCO. The product was decom-
posed by boiling water at 4-6 atm., with formation of ammonia :
2A1N + 6H2O = 2A1(OH)3 + 2NH3.

Preparation of ammonia in the laboratory. — In the laboratory,
ammonia gas is prepared by heating ammonium chloride or sulphate
with dry slaked lime :

2NH4C1 + Ca(OH)2 = CaCl2 -f 2NH3 -f 2H20.

EXPT. 205. — Mix 50 gm. of powdered ammonium chloride with 150
gm. of powdered slaked lime in a mortar, transfer to a 250 c.c. flask,
and fill up the latter with small lumps of quicklime. Fit a cork and
delivery tube, leading
to a drying tower
filled with lumps of
quicklime or caustic
soda, heat the flask
on wire gauze, and
collect the gas by
upward displacement
(Fig. 286), or over
mercury. The jar is
full when a piece of
moist red litmus paper
held near the mouth
is turned strongly
blue. After drying
with caustic
soda or potash,
the gas may be
dried with
phosphorus
pentoxide.
Concentrated
sulphuric acid

reacts violently with the gas, forming ammonium sulphate, (NH4)2SO4,
and calcium chloride absorbs it, forming a compound, CaCl2,8NH3 ;
hence these reagents cannot be used to dry ammonia.

Ammonia is also produced by heating ammonium sulphate :
(NH4)2S04 = NH3 + NH4HS04 ; microcosmic salt : NH4HNaPO4
= NH3 -j- H20 -f- NaP03 (sodium metaphosphate) ; or ammonium
phosphate : (NH4)3P04 = 3NH3 + H2O -f HP03. It is also formed
when ammonium salts are heated with a concentrated solution of
caustic soda : (NH4)2S04 -f 2NaOH = Na2S04 -f 2H2O -f 2NH3,
or when ammonium chloride is heated with litharge : PbO -f- NH4C1
^ Pb(OH)Cl -f- NH3. A regular stream of gas is evolved on

N N

FIG. 286. — Preparation of Ammonia Gas.

546 INORGANIC CHEMISTRY CHAP.

warming 170 gm. of ammonium sulphate with 250 c.c. of 50 per
cent, caustic soda solution.

EXPT. 206. — The most convenient method is to warm the concentrated
aqueous solution (liquor ammonias fortis, sp. gr. 0-88), alone or after
saturation with fused calcium chloride, in a flask ; the gas is dried with
quicklime. The solution may also be dropped on lumps of caustic
soda.

Ammonia is formed by the reduction of oxygen compounds
of nitrogen with nascent hydrogen. Thus, if a mixture of hydrogen
and nitric oxide (or a higher oxide of nitrogen, or even nitric acid
vapour) is passed over heated spongy platinum, ammonia is pro-
duced : 2NO -|- 5H2 = 2NH? + 2H2O. Dilute nitric acid in pre-
sence of dilute sulphuric acid is reduced by zinc to ammonium
sulphate : HN03 + 8H = NH3 -f- 3H20. Sodium nitrate, or more
readily sodium nitrite, is reduced by zinc and hot caustic soda
solution, giving pure ammonia. Aluminium may be used instead
of zinc, but nitrates are most easily reduced in alkaline solution
by powdered Devarda's alloy, containing 45 parts of Al, 50 parts
of Cu, and 5 parts of zinc.

EXPT. 207. — Dissolve 10 gm. of sodium nitrite and 10 gm. of caustic
soda in 50 c.c. of water and heat with a fow pieces of granulated zinc
in a flask. Ammonia is given off, turning red litmus paper blue.
This method is used for the estimation of nitrates or nitrites, the
ammonia being distilled into standard acid.

Properties of ammonia. — Ammonia is a colourless gas, lighter
than air (sp. gr. 0-59, air = 1), normal density 0-7708 gm./lit. It is
easily liquefied by cold or pressure, forming a colourless liquid,
b.-pt. — 33-5°, freezing to an ice-like solid, m.-pt. — 77°. The
critical temperature is 130°, and the critical pressure 115 atm.
The liquid may be obtained by cooling with a mixture of ice and
crystalline calcium chloride ; it is produced on a large scale by com-
pressing the gas into steel coils cooled with water, and is sent out
in steel cylinders holding 25, 50, or 100 Ib. (anhydrous ammonia).
The gas has a characteristic pungent smell, and is readily soluble
in water. The solution is alkaline.

EXPT. 208. — Fit a round-bottom flask full of ammonia gas with a
cork and tube dipping into water coloured with red litmus. Proceed
as in EXPT. 103 : the water rushes in as a fountain, and the litmus is
turned blue (Fig. 127).

The aqueous solution is prepared by passing the gas into cold
distilled water ; the flask must be kept cool by running water
over the outside from a perforated ring of lead pipe (Fig. 126),
since a considerable amount of heat is evolved. The liquid also

xxvin NITROGEN AND ITS COMPOUNDS 547

expands considerably. The saturated solution has a sp. gr. of
0-884 and contains 36 per cent, of NH3 :

Per cent. Per cent.

Sp. gr. of NH3 Sp. gr. of NH3

0-8844 36-0 0-9251 20-0

0-8864 35-0 0-9414 15-0

0-8976 30-0 0-9593 10-0

0-9106 25-0 0-9790 5-0

The aqueous solution is alkaline : it contains ammonium hydroxide,
together with much free ammonia : NH3 -f- H20 ±=; NH4OH;=r
NH4' -f OH'- By strong cooling, the crystalline hydrates NH3,H20,
or NH4OH, ammonium hydroxide, m.-pt. ~ 79-3°, and 2NH3,H2O
or (NH4)20, ammonium oxide, m.-pt. -78-6°, are obtained. A
crystalline ammonium peroxide, (NH4)202, is formed by the action
of ammonia on cold concentrated hydrogen peroxide.

Ammonia is soluble in alcohol : 1 litre of alcohol dissolves 130 gm.
of NH3 at 0°. The solubility of ammonia in water obeys Henry's
law only above 100° : all the gas is expelled on boiling a solution.
Since a considerable amount of heat is evolved on solution of the
gas, there is a large fall of temperature when the gas is removed
from the solution by a stream of air.

EXPT. 209. — Pass a rapid stream of air from bellows through a little
concentrated ammonia in a small flask standing on a wetted block of
wood. The flask is frozen firmly to the block. A temperature of
— 40° (at which mercury freezes) can be reached by rapid evaporation.
The method has been applied in some ice machines (Carre's).

The evaporation of liquid ammonia (not the solution) in steel
pipes is used in freezing machines (p. 202). The gas produced is again
liquefied by compression into steel coils immersed in cold water.

If ammonia is passed over heated potassium or sodium, one-third
of the hydrogen is replaced by the metal, and potassamide, KNH2,
or sodamide, NaNH2, is formed. These are white solids when pure.
They contain the univalent amino-group, NH2— .

EXPT. 210. — Pass ammonia, dried over quicklime or caustic potash,
over a piece of potassium heated in a hard glass bulb tube. The metal
boils, emitting a green vapour, and reaction then begins. The hydrogen
evolved may be kindled at the end of the tube, and a brown mass of
impure potassamide is left in the tube.

The compounds are violently decomposed by water, with evolu-
tion of ammonia : NaNH2 + HOH = NaOH + NH3.

Ammonia is not combustible, and does not support combustion,
but the flame of a taper, before it is extinguished in the gas, is
surrounded by a large greenish-yellow flame, due to decomposition

N N 2

548 INORGANIC CHEMISTRY CHAP.

of ammonia by heat : 2NH3 = N2 -j- 3H2. Ammonia burns in
oxygen with a greenish-yellow flame : 4NH3 -f- 302 = 6H2O -j- 2N2.
The gas is first decomposed to a large extent into its elements.

rt EXPT. 211. — Pass a current of ammonia

/,'J\ through a tube surrounded by a wider tube

through which oxygen gas is passing (Fig. 287).
If a taper is held over the tubes, the ammonia
burns with a large, double- cone, yellowish
flame.

EXPT. 212. — Prepare mixtures of oxygen
and ammonia gas over mercury in four strong
glass tubes, 8 in. long and 1 in. diameter, sealed
at one end, in the proportions : 4NH3 -f- 3O2 ;

4NH + 50

NH + 5O

2NH

13O

Ignite the mixtures with a taper. The first
two burn with violent explosions ; the remain-
ing mixtures explode less violently, and with
excess of oxygen a little red fume of NO2 and
white fumes of ammonium nitrite and nitrate are formed.

FIG. 287.— Combustion of
Ammonia in Oxygen.

EXPT. 213. — Pass oxygen through a little concentrated ammonia
warmed in a 100 c.c. conical flask, and suspend a red-hot spiral of
platinum wire in the flask. The mixture of ammonia and oxygen
explodes feebly : 4NH3 + 3O2 = 6H2O + 2N2. The wire cools, owing
to combustion ceasing, but after a short time there is another ex-
plosion, when the gas mixture is renewed. During oxidation without
explosion, red fumes of oxides of nitrogen and white fumes of
ammonium nitrate are formed :

4NH3 + 502 = 4NO + 6H2O ; 2NO + O2 = 2NO2 ;
4N02 + 02 + 2H20 + 4NH3 = 4NH4-NO3.

Ammonia is readily absorbed by dry silver chloride, forming
the compounds AgCl,3NH3 below 15°, and 2AgCl,3NH3 above 20°.
If the compound is sealed up in one limb of a bent tube (Fig. 124)
and gently heated, liquid ammonia collects in the other limb,
immersed in a freezing mixture. On allowing the silver chloride
to cool the ammonia is reabsorbed.

Ammonia is not easily decomposed by heat, especially if diluted
with an indifferent gas. A mixture of ammonia and air may also
be passed through an iron tube heated to dull redness without
appreciable decomposition.

The composition of ammonia. — If electric sparks are passed for
some time through ammonia gas in a eudiometer, it will be found
that the volume is nearly doubled. If oxygen is now added and a

NITROGEN AND ITS COMPOUNDS

549

volume of nitrogen
3 vols. of hydrogen

XXVIII

spark passed, water is formed, and two-thirds of the contraction
is equal to the volume of the hydrogen. E.g.,

Volume of ammonia taken = 20 c.c.

Volume of gas after sparking = 40 c.c.

Volume after addition of oxygen = 120 c.c.

Volume after explosion = 75 c.c.

.'. contraction on explosion with oxygen = 120 — 75 = 45 c.c.
.'. volume of hydrogen = f X 45 = 30 c.c.
= 40 — 30 = 10 c.c. Thus 1 vol. of nitrogen
= 2 vols. of ammonia (Henry, 1809).

If a concentrated solution
of ammonia containing a
little ammonium sulphate
(not chloride, as explosive
nitrogen chloride may be
formed) is electrolysed, 1
vol. of nitrogen collects at
the anode to 3 vols. of
hydrogen at the cathode.

EXPT. 214. — A long tube
(Fig. 288) is divided below
the stopcock into three equal
volumes by rubber bands,
and is filled with chlorine.
The tube above the stopcock
is two-thirds filled with con-
centrated ammonia solution,
which is added drop by drop
to the chlorine. Each drop
reacts with a
yellowish -green
flame, and the
formation of white
clouds of am-
monium chloride
(2NH3 + 3C12 =
6HC1 + N2 ; HC1 +
NH3 = NH4C1 ; cf.
p. 555). The fumes
are washed down FIG. 288.— Volumetric Composition of Ammonia (Hof^pnn).
by shaking, and

the tube is warmed in hot water to expel the nitrogen from
the liquid. Dilute sulphuric acid is then added to fix the excess
of ammonia. The tube is cooled by immersing in a large cylinder of

550 INORGANIC CHEMISTRY CHAP.

water, and the upper part above the tap is fitted with a cork and siphon
tube dipping into previously boiled water, the whole being filled with
water, as shown. On opening the tap, water rushes into the tube, and
when the levels are equalised it is found that -the residual nitrogen
occupies 1 vol.

The 3 vols. of chlorine have combined with 3 vols. of hydrogen from
the ammonia to form HC1, /. 1 vol. of nitrogen is combined in ammonia
with 3 vols of hydrogen. (Hofmann.)

The gravimetric analysis of ammonia is performed by passing
a measured volume of dry ammonia, the weight of which under
the given conditions may be calculated from the density, over red-
hot copper oxide in a hard glass tube. The water formed is collected
in weighed calcium chloride tubes. The nitrogen passing on is
collected and measured, and its weight calculated from the density.
In this way the ratio N : H is found to be 14 : 3. This, taken in
conjunction with the density of ammonia, and the volume ratio,
gives the formula NH3. The relative density of the gas is 8-552,
corresponding with the molecular weight 17-10, i.e., approximately
(since the gas does not obey Boyle's law exactly) with NH3 =
13-897 + 3 = 16-906. The atomic weight of nitrogen has been
determined from the analysis of ammonia : N = 13-897, directly
with respect to hydrogen. The previous determinations, referred
to O = 16, involved the atomic weight of silver through the ratio :
Ag : AgN03.

By-product ammonia. — Large quantities of ammonia and ammo-
nium salts, especially ammonium sulphate, are recovered as by-
products in the manufacture of gas or coke from coal. Bituminous
coal contains about 1 per cent, of nitrogen, a portion of which is
recovered in destructive distillation (p. 680). The nitrogen then
comes over mainly in the form of ammonia, although a little hydro-
cyanic acid, HCN, is present. The ammonia combines with
sulphuretted hydrogen, carbon dioxide, and sulphur dioxide,
which are also produced, to form salts, which dissolve in the water
in the coolers and scrubbers, giving ammoniacal liquor. The average
yield of ammonia in gas-works and coke-ovens is 20-25 Ib. of ammo-
nium sulphate per ton of coal, representing less than 20 per cent,
of the nitrogen in the latter. Most of the nitrogen remains in the
coke, and a further supply of ammonia, reaching a total recovery
of 60 per cent, of the nitrogen in the fuel, may be obtained by car-
bonising the latter in a current of steam, or by blowing steam
through the coke (p. 706).

Ammoniacal liquor contains tar and organic compounds, and ammo-
nium salts of two kinds : (1) Volatile salts, expelled by hydrolysis on
boiling alone ; e.g., ammonium carbonates, sulphide and hydrosulphide,
cyanide, acetate (?), and hydroxide. (2) Fixed salts, not decomposed

xxvm NITROGEN AND ITS COMPOUNDS 551

by boiling, but decomposed by lime ; e.g., ammonium sulphate, sulphite,
thiosulphate, thiocarbonate, chloride, thiocyanate, and ferrocyanide.
The total ammonia may be about 17 gm. per litre.

The ammonia is recovered from this liquor by means of ammonia
stills, in which the liquor is heated by steam to drive out the free
ammonia, or that produced by the hydrolysis of the volatile salts

FlQ. 289.— Feldman's Ammonia Still.

and the residue is then treated with milk of lime and additional
steam to decompose the fixed salts. E.g., NHJES ^ NH3 -j- H2S ;
2NH4C1 + Ca(OH)2 = 2NH3 + CaCl2 + 2H2O.

A typical still, the Feldman still, is shown in Fig. 289. It con-
sists of two iron columns containing perforated plates. In the first
column, A, the ammoniacal liquor is treated with steam to drive
out the volatile ammonia. Milk of lime from H is then added
in the lower part, B, of this column, and the sludge allowed to

552 INORGANIC CHEMISTRY CHAP.

settle. The clarified liquor then passes to the second column, C,
where the ammonia set free by the lime is driven out by
steam.

The ammonia from the still may be bubbled through 60 per cent,
sulphuric acid in a lead-lined tank, E, when crystals of ammonium
sulphate separate ; these after draining contain 93—99 per cent,
of (NH4)2S04 with a little tarry matter and free acid. If
the gas is passed through a washer containing milk of lime, to
remove sulphuretted hydrogen, and then through charcoal, or a
heavy oil washer, to remove tarry matter, it may be dissolved in
water to form a solution. Usually " 25 per cent, liquor " is made ;
the special strong liquor of density 0-884 (35 per cent. NH3) requires
very careful cooling in its preparation. About 400,000 tons of
by-product ammonium sulphate are annually prepared in Great
Britain : it is nearly all used in agriculture as a fertiliser.

Attempts have recently been made to recover ammonia from the
crude gas, from gas-works or coke-pvens, by passing the gas through
sulphuric acid without previous deposition of ammoniacal liquor.
This direct process is working in connection with coke-ovens at Skinnin-
grove and elsewhere.

Hydroxylamine, NH2'OH. — Hydroxylamine, or hydroxy-ammonia,
NH2-OH, was discovered by Lossen in 1865. He obtained its salts
by two methods :

1. By the reduction of nitric oxide, NO, with nascent hydrogen :

NO + 3H = NH2-OH.

A stream of nitric oxide, from the action of dilute nitric acid on
copper, is passed through a series of flasks containing granulated tin,
concentrated hydrochloric acid, and a few drops oi platinic chloride.
Reduction occurs, with the formation of ammonium chloride, NH4C1,
and hydroxylamine hydrochloride, NH2-OH,HC1 [or hydroxylammonium
chloride, NH3(OH)C1]. The solution is treated with sulphuretted
hydrogen to precipitate tin as stannous and stannic sulphides, filtered,
and evaporated to dryness. The residue is extracted first with cold,
then with boiling absolute alcohol, which dissolves the hydroxylamine
salt, but not the ammonium chloride. Hydroxylamine hydrochloride
is then precipitated from the alcoholic solution by adding ether.

2. By the reduction of ethyl nitrate, C2H5N03, by nascent hydro-

gen : C2H5N03 + 6H = C2H5-OH + NH2-OH + H2O.

Thirty gm. of C2H5NO3, 120 gm. of granulated tin, and 40 gm. of
HC1 (sp. gr. r06) are mixed, when reaction occurs spontaneously. The
solution is treated as in (1). This is a convenient method of pre-
paration.

xxvm NITROGEN AND ITS COMPOUNDS 553

3. Hydroxylamine salts may be prepared by the electrolytic
reduction of nitric acid (Tafel, 1902) :

HN03 + 6H = NH2-OH + 2H2O.

A cooled lead anode is separated by a porous pot from an amal-
gamated lead beaker serving as a cathode, the whole being cooled
by ice. Fifty per cent, sulphuric acid is placed in each compartment,
and 50 per cent, nitric acid added drop by drop to the cathode com-
partment. Hydroxylamine sulphate, NH2-OH,H2SO4, is formed.

4. A very convenient method of preparing hydroxylamine salts
is by the interaction of nitrites and sulphites in solution (Raschig,
1888) :

NaN02 + NaOH + 2S02 + 2H20 = NH2-OH,H2S04 + Na2S04.

EXPT. 215. — A concentrated solution of 2 mols. of commercial
NaNO2 + 1 mol. of Na2CO3 is treated with sulphur dioxide at — 2°
till just acid, keeping well stirred. The solution now contains sodium
hydroxylamine disulphonate, HO-N(SO3Na)2, i.e., HONH2 with 2H
replaced by 2SO3Na. If the solution is warmed with a few drops of
sulphuric acid, hydrolysis occurs, and sodium hydroxylamine mono-
sulphonate, HONH(SO3Na), is formed. If kept at 90-95° for two
days, further hydrolysis occurs, with formation of hydroxylamine
sulphate, NH2-OH,H2SO4. The solution is neutralised with soda,
evaporated to a small bulk, and cooled to 0°, when Glauber's salt,
Na2SO 4, 10H2O, crystallises out. The filtrate on further evaporation
deposits hydroxylamine sulphate, which is rapidly recrystallised from
water.

The reaction occurs in three stages, as follows :

(a) NaN02 + 3NaHS03 = HON(S03Na)2 + Na2S03 + H2O

Sodium hydpoxylamine disulphonate.

(6) HO-N(S03Na)2 + H20 = HONH(S03Na) + NaHSO4

Sodium hydroxylamine monosulphonate.
(c) HO-NH(S03Na) + H2O = HONH2 + NaHS04

Hydroxylamine*

By these methods salts of hydroxylamine are produced : if caustic
potash is added to a solution of a salt, free hydroxylamine is first
formed, but is unstable : 3NH2-OH = NH3 + 3H2O + N2.

Anhydrous hydroxylamine, NH2-OH, was prepared by Lobry de
Bruyn in 1891 by treating a solution of the hydrochloride in methyl
alcohol with a solution of sodium methoxide in methyl alcohol
(obtained by dissolving sodium in the alcohol : 2CH3-OH -f 2Na =
2CH3ONa + H2), filtering off the sodium chloride, and distilling
under reduced pressure (40 mm.) : CH3ONa + NH2-OH,HC1 =
CH3-OH -f NaCl + NH2-OH. Crismer (1891) distilled the double

554 INORGANIC CHEMISTRY CHAP.

compound ZnCl2,2NH2'OH (obtained by boiling zinc oxide with a
solution of hydroxylamine hydrochloride) at 120°, either alone or
with aniline. Anhydrous hydroxylamine is also formed by heating
the orthophosphate to 135° under very low pressure (13 mm.) :
(NH40)3P04 = H3P04 + 3NH30.

Properties of hydroxylamine. — Pure hydroxylamine forms colour-
less, odourless crystals, sp. gr. 1-3, m.-pt. 33°. It is very
deliquescent. It may be distilled under reduced pressure
(55-58 °/22 mm.), but explodes when heated at the ordinary
pressure. The vapour density corresponds with NH30. Above
15° it slowly decomposes, evolving nitrogen and nitrous oxide.
Aqueous solutions containing up to 60 per cent, are fairly
stable. The vapour explodes in contact with air at 60-70°. The
solution is strongly basic and precipitates many metals (Zn, Al)
as hydroxides.

Hydroxylamine and its salts in aqueous solution act as powerful
reducing agents. "They precipitate red cuprous oxide from copper
sulphate in alkaline solution, purple metallic gold from gold chloride,
and in acid solutions reduce ferric to ferrous salts :

2NH30 -f 4CuO = N20 + 3H20 + 2Cu2O
4FeCl3 + 2NH30 = N2O + 4FeCl2 + 4HC1 + H20.

In alkaline solution, hydroxylamine oxidises ferrous hydroxide
to ferric hydroxide, with formation of ammonia :

2Fe(OH)2 + NH3O + H20 = 2Fe(OH)3 + NH3.

The salts on heating with nitric acid evolve nitric oxide :
NH?0 + HN03 = 2NO -f 2H20. When treated with a nitrite and
acidified, they evolve nitrous oxide on warming. Hyponitrous
acid, H2N202, is formed as an intermediate product :

HO-NH2 + ON -OH = HO-N:N-OH + H20 = N20 -f 2H20.

Nitrous acid Hyponitrous acid

Hydroxylamine reacts with organic substances containing the
aldehyde, — COH, or ketone >CO, groups, forming oximes :

H2N-OH - UN>C:N-OH + H2O.
B/

On hydrolysis, these give hydroxylamine. Fulminic acid, C:N-OH,
on boiling with hydrochloric acid gives hydroxylamine. If a neutral
solution containing a hydroxylamine salt is treated with sodium nitro-
prusside, and a little caustic soda, a beautiful red colour appears on
boiling (test).

Nitrogen trichloride. — Dulong (1811) by the action of chlorine
on a solution of ammonium chloride obtained a yellow oily liquid
which was violently explosive. He lost an eye and a finger in the
research. Davy and Faraday (1813) obtained the compound by

NITROGEN AND ITS COMPOUNDS

555

the action of excess of chlorine on ammonia, and concluded that its
formula was NC14. Balard prepared it by the action of hypochlorous
acid on ammonia, and B.ottger and Kolbe found that it separated
at the anode in the electrolysis of ammonium chloride solution at
28°. The substance is nitrogen trichloride, NC13 :

3C12 + 3H2O =± 3HOC1 + 3HC1
NH4C1 + 3HOC1 =^ NC13 + 3H20 + HC1.

Nitrogen trichloride may be prepared by inverting a flask of chlorine
over a 25 per cent, freshly pre-
pared solution of ammonium
chloride, a lead saucer being
placed under the mouth of the
flask (Fig. 290). The chlorine
is absorbed, and oily drops of
the trichloride float on the
surface of the solution. These
fall into the lead saucer, which
should be removed when a little
liquid has collected in it. If a
little turpentine is passed into
the flask, a violent explosion
results, the glass being com-
pletely shattered. The drop
of oil in the dish also explodes
violently when touched with
a feather dipped in turpen-

tine. This experiment should Fm> 290._preparation of Nitrogen Trichloride.

be tried in the open air,

and with adequate precautions, only by an experienced chemist.

EXPT. 216. — The formation of nitrogen chloride may be safely shown
by the apparatus of Fig. 291. The solution
of ammonium chloride saturated at 28°
is poured into the tube, closed at the
lower end with a piece of moist bladder,
and the whole dipped into a trough of the
solution. Electrodes of platinum foil are
immersed in the tube and dish, and a
layer of turpentine is poured over the solu-
tion in the tube. The electrode in the
tube is made the anode. As each drop
of trichloride rises and enters the turpen-
tine it explodes, forming nitrogen and chlorine.

Gattermann found that the trichloride had the formula NCI a

FIG. 291. — Demonstration of
Explosion of NCla.

556 INORGANIC CHEMISTRY CHAP.

if the action of chlorine was prolonged, but the chlorination of
ammonia proceeds in three stages :

NH3 + C12 = NH2C1 (monochloramine) + HC1 ;
NH2C1 + C12 = NHC12 (dichloramine) + HC1 ;
NHC12 + C12 = NC13 (trichloramine) + HC1.

The analysis was carried out by decomposing with ammonia :
NC13 + 4NH3 = N2 + 3NH4C1, and precipitating the chloride with
silver nitrate. The percentage of chlorine was found to be 89-1 ;
NC13 requires 89 -17.

Monochloroamine, NH2C1, is formed as an unstable yellow liquid
when ammonia and sodium hypochlorite are mixed in equimolecular
proportions and the liquid is distilled in a vacuum : NaOCl + NH3 =
NaOH + NH2C1. By the action of nitrogen trichloride on potassium
bromide, Millon obtained a dark red, volatile, explosive oil, possibly
nitrogen tribromide.

Nitrogen iodide. — By the action of iodine on a solution of ammonia,
Courtois (1829) obtained a black explosive powder. This was
examined by Gladstone (1855), who gave it the formula NHI2,
whilst Stahlschmidt (1863) considered it to be NI3. Bunsen (1852),
by mixing alcoholic solutions of iodine and ammonia, obtained
N2I3H3, i.e., NI3-NH3. Szuhay (1893), by suspending the black
" iodide of nitrogen " in water and adding silver nitrate, obtained
a black explosive powder, which he stated to have the composition
NAgI2. The formula NHI2 was therefore considered to be correct.
Chattaway and Or ton (1900) found, however, that the first product
of the action of iodine on aqueous ammonia is a dark red crystalline
compound, NI3-NH3, and they confirmed the observation of
Selivanoff (1894) that hypoiodous acid is the first product of the
reaction. This appears to react with more ammonia to form the
iodide of nitrogen, possibly by decomposition of ammonium hypo-
iodite : (a) NH4-OH + I2 = NH4I -f HOI : (6) NH3 + HOI =
NH4OI ; (c) 3NH4OI .^± N2H3I3 + NH4-OH + 2H20.

If iodide of nitrogen is treated with sodium sulphite it is decom-
posed : N2H3I3 + 3Na2S03 + 3H20 = 2Na2S04 + 2NH4I + HI.
The free acid may be titrated with baryta, and the iodide with silver
nitrate, and the composition of the substance so determined.
Silberrad (1905) confirmed the formula by the action of zinc ethyl
on the substance :

N2H3I3 + 3Zn(C2H5)2 = 3ZnC2H5I + NH8 + N(C2H5)3.
He showed that Szuhay's compound is NI3-AgNH2.

EXPT. 217. — If a dilute solution of iodine is added drop by drop to a
solution of ammonia, the liquid at first remains clear, and gives the
reactions of hypoiodous acid (e.g., a brown precipitate with MnSO4).

xxvin NITROGEN AND ITS COMPOUNDS 557

On further addition of iodine, a black precipitate of iodide of nitrogen
is formed. If a large amount of concentrated ammonia is added, this
redissolves, showing that reaction (c) above is reversible.

EXPT. 218. — Triturate 1 gm. of iodine with concentrated ammonia.
A black powder is formed, which is filtered off. The iodide of nitrogen
is fairly stable when moist. The filter-paper is torn into a number of
pieces, which are allowed to dry spontaneously. If one portion is
touched with a feather, it explodes — sometimes spontaneous explosion
occurs. If the other portions are not kept at a distance they also
explode. If two portions are placed close together, and one is exploded,
the shock brings about the explosion of the other portion. Violet
fumes of iodine are evolved. If one of the portions of the moist sub-
stance is placed in water and exposed to light, bubbles of nitrogen are
evolved. Another portion of the moist iodide may be dissolved in
concentrated ammonia, when a brown solution containing iodine and
ammonium iodate is formed on warming.

Nitrogen iodide is an active oxidising agent, oxidising sulphites
to sulphates, arsenious acid into arsenic acid, etc. Each atom of
iodine has an oxidising effect of an atom of oxygen, as in hypoiodous
acid, HOI.

Hydrazine, N2H-4.— Hydrazine, or diamide, N2H4, was prepared
by Curtius in 1887 from organic compounds. Raschig obtained it
by the action of sodium hypochlorite on ammonia solution in the
presence of a little glue. Monochloroamine is first formed, which
then reacts with ammonia to form hydrazine :

NH3 + NaOCl = NH2C1

NH3-fNH2Cl = NH2-NH2

EXPT. 219.- — One litre of commercial sodium hypochlorite solution

is mixed with 12 c.c. of a 5 per cent, solution of glue and added to 3

litres of concentrated ammonia. The solution is concentrated by

evaporation to drive off excess of ammonia, and neutralised with

sulphuric acid. On cooling, 80-90 gm. of hydrazine sulphate,

N2H4,H2SO4, are obtained.

Hydrazine is also formed by the reduction of potassium nitroso-
hydroxylamine sulphonate, obtained by saturating a solution of potassium
nitrite with sulphur dioxide. This salt, which has the empirical
formula K2SO3-N2O2, is suspended in ice-cold water and treated with
sodium amalgam :

KSO3V KSO3.

>N-NO + 6H = >N-NH2 + H2O + KOH

KO/ H/

TC^O

3\N-NH2 + KOH = K2S04 -

558 INORGANIC CHEMISTRY CHAP.

If hydrazine sulphate is distilled under reduced pressure with
concentrated potash solution, with a condenser without rubber or
cork connections, a colourless fuming liquid, b.-pt. 119°, or
47°/26 mm., is obtained. This is called hydyazine hydrate, N2H4,H20,
but appears to be a solution of maximum boiling point (p. 237).
If the hydrate is distilled with its own weight of caustic soda in small
pieces, anhydrous hydrazine passes over at 150° as a liquid which
solidifies on cooling into colourless crystals, m.-pt. 1 4°, b.-pt. 113-5°.
Anhydrous hydrazine may also be prepared from the hydrochloride
and sodium methoxide (p. 553). Hydrazine, and the hydrate,
readily absorb moisture and carbon dioxide from the air, are freely
soluble in water and alcohol, and are poisonous. Anhydrous
hydrazine inflames in dry oxygen, reacts readily with halogens :
2I2 -f- N2H4 = 4HI -(- N2, explodes in contact with potassium per-
manganate, sets free ammonia from ammonium chloride, and
decomposes on heating : 3N2H4 = N2 -f 4NH3.

Hydrazine in solution acts as a very weak base : it forms two
series of salts, e.g., N2H4,HC1, N2H4,2HC1 ; 2N2H4,H2S04,
N2H4,H2SO4. The ordinary hydrazine sulphate is N2H4,H2S04, or
(N2H6)HS04. The salts are ionised and hydrolysecl in solution :
N2H4,2HX ^± N2H4,HX + HX — N2H5' + H' + 2X'. Double
salts, e.g., ZnCl2,N2H4,2HCl, are known.

Hydrazine and its salts are the most powerful reducing agents
known, precipitating gold, silver, and platinum from their
salts, reducing alkaline copper solutions to cuprous oxide :
4CuO + N2H4 = 2Cu2O + 2H2O -f N2 ; ferric salts to ferrous
salts, and iodates to iodides : 3N2H4,H2S04 + 2KI03 =
2HI + 2KHS04 + H2S04 -f 6H20 -f 3N2. Hydrazine may be esti-
mated by titration with iodine in presence of sodium bicarbonate :
N2H4 -f- 2I2 = N2 -f- 4HI, or with potassium permanganate in pre-
sence of dilute sulphuric acid : N2H4 -f 20 = N2 -f- 2H20.

Hydra zoic acid, HN3. — Hydrazoic acid, or azoimide, HN3, was
obtained by Curtius in 1890 from organic compounds. It is formed
by the careful oxidation of hydrazine with nitric acid or hydrogen
peroxide : 3N2H4 + 30 = 2HN3 + 3H20.

EXPT. 220. — Warm 1 gm. of hydrazine sulphate with 4 c.c. of HNO3
of sp. gr. 1-3 in a test-tube, and lead the vapours into a solution of
silver nitrate. A white, curdy precipitate of silver azide, AgN3, is
formed. This compound is explosive when dry. It is soluble in
ammonia (cf. AgCl).

Hydrazoic acid is also formed by the decomposition of hydrazine
nitrite under special conditions :

N2H4,HN02 = HN-g 4 2H20
(cf. NH3,HN02 =N2 42H20).

xxvni NITROGEN AND ITS COMPOUNDS 559

If hydrazine is treated with ethyl or amyl nitrite and alkali,
sodium azide is formed, and a precipitate of silver azide is pro-
duced when hydrazine is added to a concentrated solution of
silver nitrite.

Wislicenus (1892) first prepared hydrazoic acid from inorganic
materials. Sodamide, NaNH2, is prepared by passing dry ammonia
over pieces of sodium in porcelain boats in a hard glass tube heated
to 150-250° : 2Na + 2NH3 = 2NaNH2 + H2. The ammonia is
then displaced by a current of dry nitrous oxide, and the tube
heated to 190°. The sodamide swells up, and ammonia is evolved :
(a) NaNH2 + N20 = NaN3 + H20 ; (6) NaNH2 + H2O =
NaOH -f NH3. When no more ammonia is evolved, the tube is
cooled, and the pumice-like mass of NaN3 and NaOH distilled with
dilute sulphuric acid, when a solution of hydrazoic acid, HN3,
comes over.

The solution is fractionated, and finally distilled with fused
calcium chloride, when anhydrous hydrazoic acid is formed. This
is a colourless mobile liquid, b.-pt. 37°, m.-pt. -- 80°, with a
nauseous smell. It is dangerously explosive. It dissolves readily
in water, forming a corrosive acid liquid, in which about 1 per cent.
of the acid is ionised : HN3 ±^ H' -f- N3'. The solution readily
dissolves iron, zinc, copper, and aluminium, with evolution
of hydrogen and ammonia : 2HN3 -f- Zn = Zn(N3)2 + H2 ;
HN3 + 6H = NH3 -f N2H4.

The salts give a blood-red colour with ferric chloride, resembling
thiocyanates ; with silver nitrate they give a white, curdy precipitate
of silver azide, AgN3, soluble in ammonia, and exploding at 250°.
By neutralising the acid with ammonia and hydrazine, respectively,
the salts NH4N3(N4H4) and N2H4-HN3(N5H6) are obtained in
colourless explosive crystals.

The constitutional formula of hydrazoic acid was formerly written

/N
as : H— N< || . Thiele represents it as N : N:NH, which agrees with

the formation of NH3 and N2H4 on reduction. The group N3 is
& negative group (cf. p. 517), whilst NH2 is a positive group ; H — NH2
is a base.

EXERCISES ON CHAPTER XXVIII

1. How is nitrogen obtained : (a) from air, (b) from ammonia,
(c) from ammonium nitrite ? How do these varieties of nitrogen differ
from one another, and what is the cause of the difference ?

2. How is pure nitrogen obtained ? What are its physical and
chemical properties?

3. With what substances, and under what conditions, does nitrogen

560 INORGANIC CHEMISTRY CH. xxvm

unite directly ? How may ammonia be prepared from atmospheric
nitrogen ?

4. What are the sources of commercial ammonia ? A specimen of
ammonia contains ammonium sulphide : how would you prepare
pure ammonia from it ?

5. How is ammonia made : (a) in the laboratory ; (b) on the large
scale ? Describe the principal properties of the gas.

6. How is the composition of ammonia determined ? Twenty c.c.
of a mixture of ammonia and nitrogen are exploded with oxygen. The
contraction is 7-5 c.c. What volumes of the gases are present in the
mixture ?

7. How is ammonium sulphate obtained on the large scale ?

8. How are hydroxylamine salts obtained ? How is hydroxylamine
obtained from its salts ?

9. What are the principal properties of hydroxylamine ? Describe
reactions in which it functions : (a) as a base ; (b) as a reducing agent ;
(c) as an oxidising agent.

10. How is hydrazine sulphate obtained ? Starting with this salt,
how would you prepare anhydrous hydrazine ? Describe the properties
of these substances .

11. What halogen compounds of nitrogen are known ? Describe
briefly their preparation and properties.

12. How is hydrazoic acid obtained ? What are the properties of
this substance ?

13. What is the action of ammonia on : (a) chlorine, (b) potassium,
(c) nitrous acid, (d) oxygen ?

CHAPTER XXIX

THE OXIDES AND OXY-ACIDS OF NITROGEN

Oxides and oxy-acids of nitrogen. — A number of oxides and oxy-
acids of nitrogen are known : the following table may be compared
with that of the oxy-compounds of chlorine (p. 373) :

Nitrous oxide, N2O -> Hyponitrous acid (N- OH) 2 or H2N202

Nitric oxide, NO ~> Nitrohydroxylamic acid, HON:N02H

Nitrogen trioxide or ) ^ Q HNO

Nitrous anhydride j
Nitrogen dioxide, or tetroxide, N02,N204 \^

T205 -> Nitric acid, HN03

Nitrogen heptoxide, N2O7 (?) -> Pernitric acid, HNO4 (?)
In many of its oxy-compounds nitrogen is quinquevalent.

TYPE NX3.
NH3, ammonia
NH2(OH), hydroxylamine
NH(OH)2, dihydroxyammonia,

or dihydroxylamine
N(OH)3, orthonitrous acid

Dehydration products : —
NH(OH)2 - H2O -> N-OH
Nitrous oxide 4

Nf\ I TT (~\ ^ I XT • /\TT \
2U -J- -TL2w ^— (a Unj.j

hyponitrous acid

TYPE NX5.

NH4 ; NH4C1, ammonium chloride
NH,-OH, ammonium hydroxide
NH3(OH)2

NH2(OH)3
NH(OH)4
N(OH)5, orthonitric acid

N(OH)3 - H20 -> NO-OH,

nitrous acid
2NO-OH -H?0 ->N203,

nitrous anhydride

Hydrazine, H2N-NH2

561

Dehydration products : —
NR3(OH)2 - H20 -» ONR3, oxy-
amines
NR(OH)4 - 2H2O-> R-N02,

nitro-compounds
N(OH)6 - H20 -> NO-(pH)8, §

rneso-nitric acid
NO(OH)3 - H2O -> NO2-OH,

nitric acid (meta)
2N02-OH - H20 -> N02-0-N02
or N205, nitric anhydride

Dehydration products : —
(HO)2NH-N(OH)3 - 2H20 ->
HON:N02H, nitrohydroxylamic

acid

HON:N02H - H2O -> 2ND,
nitric oxide.
O O

562

INORGANIC CHEMISTRY

CHAP.

Compounds in heavy type are known ; others are hypothetical.
R indicates an organic radical, e.g., ethyl, C2H5 — .

The union of nitrogen and oxygen. — Nitrogen and oxygen combine
directly at high temperatures to form nitric oxide: N2 + 02^±
2NO. With excess of oxygen this forms on cooling red fumes of
nitrogen dioxide : 2NO + O2 = 2NO2. If water is present as well
as excess of oxygen, the nitrogen dioxide dissolves, forming a
solution of nitrous and nitric acids : 2N02 + H2O = HNO2 -f HNO3.
Nitrous acid is unstable, the solution becoming pale blue in colour
owing to the formation of nitrous anhydride, N2O3. This also
decomposes, forming nitrogen dioxide and nitric oxide, which are

FIG. 292.— Combination of Nitrogen and Oxygen by Sparking.

evolved : N2O3^N02 + NO. The nitric oxide is again oxidised,
and finally all the oxides of nitrogen are converted into dilute
nitric acid.

EXPT. 221. — Pass a series of sparks through air in a globe (Fig. 292).
After a time the gas becomes yellowish in colour, and if it is shaken
with litmus solution the latter is turned red. This observation was
first made by Priestley (1779).

Nitric acid is also formed when a mixture of detonating gas
(2 vols. of H2 -{- 1 vol. of O2) with air is exploded by a spark. If
the volume of air is more than double that of the detonating gas,
the temperature of explosion is too low to lead to the formation
of nitric acid. Thus, no acid is formed on exploding a mixture
of hydrogen and air. This observation is due to Cavendish
(1781).

xxix THE OXIDES AND OXY-ACIDS OF NITROGEN 563

If nitric acid is distilled with phosphorus pentoxide, nitric
anhydride is formed : 2HN03 = N205 + H20. By the action of
dilute nitric acid on zinc, or by heating ammonium nitrate, nitrous
oxide is obtained : NH4N03 = N2O -f 2H2O.

Nitre or saltpetre. — If soil containing decomposing nitrogenous
organic matter, such as urine, is mixed with lime or calcium car-
bonate, such as old mortar, calcium nitrate, Ca(N03)2, is produced.
It probably arises from the oxidation of ammonia, formed by the
decomposition of organic matter, in the presence of feeble alkalies,
by the activity of micro-organisms, known as nitrifying bacteria,
present in all fertile soil. If an infusion of soil is added to a dilute
solution of an ammonium salt containing calcium carbonate in
suspension, calcium nitrate is formed. The first product of oxida-
tion may be calcium nitrite, which is then fully oxidised to nitrate,
and two kinds of bacteria are usually involved :

+ 3O +O

NH3 -> HN02 -> HN03.

There are, however, bacteria which convert ammonium salts
directly into nitrates. If the material is lixiviated, a solution
containing calcium nitrate is obtained, which is boiled with wood-
ashes (containing potassium carbonate) : Ca(N03)2 + K2CO3 =
CaCO3 -j- 2KN03. The filtrate on evaporation deposits prismatic
crystals of nitre, saltpetre, or potassium nitrate, KN03.

This method of obtaining nitre by means of nitre plantations is
still used in India, where about 20,000 tons are made annually.

Potassium nitrate usually crystallises in large rhombic prisms,
but if the solution is slowly evaporated on a watch-glass, rhombo-
hedra, isomorphous with sodium nitrate, deposit. The rhombic
form is stable below 129°, a second rhombohedral form at higher
temperatures. Nitre melts at 339°, and the fused salt is a power-
ful oxidising agent. Sulphur, charcoal, and phosphorus take fire
in it, and burn brilliantly, with formation of potassium sulphate,
carbonate, and phosphate. This property is applied in the
manufacture of gunpowder.

Potassium nitrate is used in pickling meat, to which it imparts
a bright red colour (e.g., hams), and in medicine. It is used as a
fertiliser, since both potassium and nitrates are essential to the
growth of plants (p. 696).

Sodium nitrate ; Chile nitre, NaN03. — In 1830 the existence of
extensive deposits of sodium nitrate, NaNO3, was discovered in the
rainless districts of Chile. The zone of nitrates appears to cover
77,000 square miles, of which less than 3 per cent, is explored and pros-
pected. In the surveyed area the supply is 240,000,000, tons, esti-
mated, with normal production, as sufficient for one hundred years.
Other authorities give three hundred years as the probable period of

o o 2

564 INORGANIC CHEMISTRY CHAP.

exhaustion of the nitre beds. The exports of sodium nitrate from
Chile have been as follows :

Year. Export in tons. Year. Export in tons.

1830-34 . . 16,780 1890 . . . . 1,000,000

1865 .. .. 491,100 1895 .. .. 1,267,000

1875 . . . . 334,000 1905 . . . . 1,705,000

1885 .. .. 512,600 1915 .. .. 2,090,000

The sodium nitrate in the deposits constitutes from 20 to 50 per
cent, in a distinct stratum of earth known as caliche, resting upon
soft clay, and covered with a compact top layer called costra,
containing less nitrate. The surface soil having been removed,
holes are bored through the costra into the caliche, charges of
slow-burning powder are inserted and tamped, and the caliche is
then broken up by the explosion. The pieces of caliche are con-
veyed to the lixiviation works, known as qfficina, where the material
is crushed and lixiviated in large tanks of water heated by steam.
The settled solution is run off to crystallisers, where crude nitrate
separates, the mother liquors being run back to the lixiviators.
The crystals are washed with a little water and dried in the sun :
they contain 95-96 per cent, of NaN03, and are exported in

About four-fifths of the export of Chile nitre is used directly as a
fertiliser: the remainder is used as a source of nitric acid, for the
manufacture of explosives, dyes, and drugs.

Sodium nitrate crystallises in rhombohedra resembling cubes,
hence it is sometimes called " cubic nitre." It differs from
potassium nitrate in being deliquescent ; it fuses at 316°, and at
higher temperatures evolves oxygen, leaving nitrite : 2NaN03 =
2NaN02 + 02.

Sodium nitrate is converted into potassium nitrate by dissolving
potassium chloride in hot water till the sp. gr. is 1-2, and then
adding sodium nitrate till the sp. gr. rises to 1-5. Sodium chloride,
the least soluble salt formed from the four ions, is deposited from
the hot liquid, since its solubility is not appreciably increased by
rise of temperature ; if the mother liquor is allowed to cool,
potassium nitrate crystallises out, since it is the least soluble salt
at lower temperatures : NaN03 + KC1 ^± KN03 + NaCl. It is
recrystallised from water.

Gunpowder. — Most of the potassium nitrate of commerce is used
in making gunpowder. This was apparently first made by the
Chinese for the production of fireworks. Greek fire was a mixture
of nitre, pitch, and sulphur. The invention of gunpowder is
usually credited to Roger Bacon (1214-1294), although it is explicitly
described by Marcus Graecus (eighth century), who also gives recipes

xxrx THE OXIDES AND OXY-ACIDS OF NITROGEN 565

for " liquid fire " for military purposes. Gunpowder was first used
by the English, in the battle of Crecy, in 1346. It consists of a
mixture of finely-powdered nitre, wood-charcoal (carbonised at a
low temperature), and sulphur, usually in the proportions
74-9 : 13-3 : 11-8, the materials being ground and incorporated
under stone rollers. (Marcus Graecus gives 60 : 20 : 10.) The
proportions of the constituents, and the main products of com-
bustion, correspond roughly with the following equation :
2KN03 + S + 3C = K2S + N2 + 3C02.

Carbon monoxide, however, is also evolved, and the residue
contains potassium carbonate and sulphate. Abel and Noble
(1875) found that the explosion of gunpowder cannot adequately
be represented by a chemical equation, since the reactions are
exceedingly complicated.

The equation given shows that (2 X 101 +32 + 3 X 12) = 270 gm.
of powder produce 4 X 22-3 litres of gas at S.T.P. The solid powder
occupies about 100 c.c., hence the expansion at S.T.P. will be about
800. The temperature of the gaseous products, at the instant of
explosion, is about 2000°, so that the theoretical liberation of energy in
firing 1 gm. of powder should be about

2273

or 2-8 X 109 ergs. This would impart to a bullet of mass 1 gm. a
muzzle velocity of V% x 2-8 x 109 = 7-5 X 104 cm. per sec.

Nitric acid, HN03. — The Latin Geber describes the preparation
of aqua fortis by distilling nitre with alum and copper sulphate :
" Take a pound of vitriol of Cyprus, a pound and a half of saltpetre,
a quarter of alum of Jameni ; submit the whole to distillation in
order to obtain a liquid which has great solvent power " (" Alchimiae
Geberi," 1529). Glauber (1603-1668) obtained a more concentrated
fuming acid by distilling nitre with oil of vitriol (" Philosophische
Oefen," 1648). The acid was
therefore known as spiritus
nitri fumans Glauberi. The
presence of oxygen in nitric
acid was demonstrated by
Lavoisier in 1776.

EXPT. 222. — Arrange a clay
tobacco pipe as shown in Fig.
293. Heat one part of the FIG. 293.— Decomposition of Nitric Acid by Heat.
stem strongly with a Bunsen
burner, and pour 5 c.c. of concentrated nitric acid into the bowl. The

566

INORGANIC CHEMISTRY

CHAP.

acid is decomposed on passing through the hot tube, and bubbles of
oxygen collect in the test-tube.

The composition of nitric acid was elucidated by Cavendish
(1784). He passed a series of sparks through a mixture of oxygen
and nitrogen confined over mercury and potash solution in an
inverted U-tube (Fig. 294). The gas gradually disappeared, with
the exception of a very small bubble (p. 600), and a solution of
nitre was formed. Thus, nitric acid is formed from oxygen and
nitrogen in the presence of water.

Cavendish says : " We may safely conclude that in the present
experiments the phlogisticated air [N] was enabled, by means of the
electric spark, to unite to, or form a chemical combination with, the

dephlogisticated air [O],
and was thereby reduced
to nitrous [nitric] acid,
which united with the
soap -lees [potash] and
formed a solution of nitre ;
for in these experiments
those two airs actually dis-
appeared, and nitrous acid
was formed injtheir room."

Nitric acid is prepared
in the laboratory by dis-
tilling potassium or sodium
nitrate with concentrated
sulphuric acid : KNO3 +
H2S04-- KHS04+HN03.
If excess of nitre is used,

and a high temperature, further decomposition occurs, the acid

sulphate being converted into normal sulphate :

KHS04 + KN03 = K2S04 + HN03.

A glass retort is then usually cracked, and part of the acid is
decomposed, with production of red fumes of oxides of nitrogen :
4HNO3 = 4N02 + 2H2O + O2. These fumes dissolve in the acid,
colouring it yellow.

EXPT. 223. — Add 49 gm. of concentrated sulphuric acid to 50 gm. of
potassium nitrate in a stoppered retort. Heat on wire gauze, and
collect the nitric acid in a cooled receiver (Fig. 295). Notice the red
fumes at the beginning and end of the process. The residue in the
retort may be poured out into a porcelain dish, and solidifies to a white
crystalline mass of impure potassium hydrogen sulphate, KHSO4.
If a little of this is powdered, mixed with powdered KNO3, and heated

FIG. 294. — Cavendish's Apparatus for Sparking Air
over Potash Solution.

xxix THE OXIDES AND OXY-ACIDS OF NITROGEN 567

in a test-tube, white fumes of nitric acid and red fumes of NO2 are
evolved. A glowing chip inflames in the* gas, showing that oxygen is
also produced.

Pure nitric acid is obtained by redistilling on a water-bath under
reduced pressure, and passing ozonised oxygen through the dis-
tillate. It is a colourless liquid of sp. gr. 1-52. The pure acid
may also be obtained by freezing 98 per cent, acid, when colourless
crystals, m.-pt. —41-3°, separate.

The liquid acid and the vapour are slightly dissociated at the
ordinary temperature : 2HNO3±=^N205 -f H2O, and the dissocia-
tion increases with the temperature. Anhydrous HNO3 does not exist
in the liquid state. If a current of dry air is passed through the
liquid acid, the
more volatile
nitric anhydride
is removed, and
an acid of con-
stant boiling
point (about
86°) containing
98-62 per cent,
of HN03, is ob-
tained. The
acid decomposes
on distillation
under atmos-
pheric pressure.
It begins to boil
at 78-2°, with
decomposi t i o n.
When three-fourths of the acid has distilled over, the residue
contains 95-8 per cent, of HN03 ; with further distillation an
acid of maximum boiling point ( 120-5°), containing 68 per cent,
of HN03, is formed. This is also formed when weaker solutions
are distilled. This acid, although it corresponds approximately
with 2HN03,3H20, is not a definite hydrate ; Roscoe showed that,
as in the case of hydrochloric acid, the
distillate is a function of the pressure.
HN03,H20 (m.-pt. - 38°) and HN03,3H20 (m.-pt. - 18-5°), are
known.

Nitric acid vapour is decomposed by light. If a bottle half filled
with acid is exposed to light, the nitrogen dioxide formed dissolves
in the liquid and renders it yellow. The liquid in a completely
filled bottle remains colourless. The yellow acid may be rendered
colourless by warming to 60-80°. and bubbling dry air through it ;

FIG. 295.— Preparation of Nitric Acid.

composition of the
Two solid hydrates,

er cent.

Per cent.

HN03
8-99

Density.
1-250

HN03
39-82

Density.
1-450

HN03

77-28

17-11

1-300

47-49

1-500

94-09

24-84

1-350

55-79

1-510

98-10

32-36

1-400

65-30

1-520

99-67

568 INORGANIC CHEMISTRY CHAP.

or by adding a little lead dioxide, when oxides of nitrogen are con-
verted into lead nitrate, which is insoluble in the concentrated acid,
and separates with the excess of dioxide : Pb02 + N204 = Pb(N03)2.

A yellow, so-called fuming nitric acid, containing oxides of nitrogen,
is used as an oxidising agent, and is prepared by distilling nitre
and sulphuric acid with a little starch. The starch reduces a por-
tion of the nitric acid to N2O3 and N2O4, which dissolve in the acid.

Heat is evolved, and contraction occurs, when concentrated
nitric acid and water are mixed. The maximum effect occurs
with the mixture 3HN03 + H2O, although no definite hydrate of
this composition has been isolated.

The densities of mixtures of nitric acid and water at 15° are given
in the table below.

Density.
1-050
1-100
1-150
1-200

Chemical properties of nitric acid. — Nitricjicid is a strong acid,
and is largely ionised in solution: HN03^H' -f- NO3'. It is
monobasic, and forms only one series of salts, the nitrates, which
are obtained by the action of nitric acid on the metals, when oxides
of nitrogen and not hydrogen are usually evolved (p. 570), on the
oxides or hydroxides, or on the carbonates.

EXPT. 224. — Dissolve copper turnings in dilute nitric acid. Observe
that red fumes are evolved. Evaporate the blue solution, and allow
to cool. Blue deliquescent crystals of cupric nitrate, Cu(NO3)2,3H2O,
are obtained.

EXPT. 225. — Neutralise a solution of caustic potash with dilute
nitric acid, evaporate, and allow to cool. Prismatic crystals of nitre,
KNO3, separate.

EXPT. 226. — Dissolve lead carbonate in warm dilute nitric acid,
filter from excess of carbonate, and evaporate. Octahedral crystals
of lead nitrate, Pb(NO3)2, are formed.

Nitric acid also acts as an oxidising agent. Concentrated nitric
acid, when boiled with iodine, oxidises the halogen to iodic acid,
HI03. Phosphorus is oxidised to phosphoric acid, sulphur to
sulphuric acid, arsenious oxide to arsenic acid. Tin is oxidised by
concentrated nitric acid in the cold, with evolution of red fumes,
and a white residue of hydrated stannic oxide remains. Burning
charcoal burns brilliantly in the concentrated acid, and heated
sawdust is inflamed.

xxix THE OXIDES AND OXY-ACIDS OF NITROGEN 569

EXPT. 227. — Heat a little sawdust on a sand-bath until it begins to
char, and pour over it a few drops of fuming nitric acid from a test-
tube. The sawdust burns.

Oil of turpentine explodes with concentrated nitric acid, with
evolution of black clouds of carbon. Alcohol is violently oxidised,
with the production of a variety of substances, and usually with
explosion.

Sulphuretted hydrogen is not oxidised by pure nitric acid, but
in presence of nitrogen oxides it is decomposed with separation of
sulphur. Stannous chloride in hydrochloric acid is oxidised to
stannic chloride ; the nitric acid is reduced to hydroxylamine and
ammonia.

Ferrous salts reduce nitric acid to nitric oxide, NO, and this
dissolves in the excess of ferrous salt to form a black solution,
from which nitric oxide is expelled on heating. The first reaction
is : 6FeS04 -4- 2HN03 + 3H2SO4 = 3Fe2(S04)3 + 2NO + 4H2O.
This is utilised as a test for nitric acid or nitrates. In the case of
nitrates, concentrated sulphuric acid must be added before the
colour appears.

EXPT. 228. — Dissolve a few crystals of ferrous sulphate in a cold
dilute solution of potassium nitrate in a test-tube, and pour pure con-
centrated sulphuric acid (the commercial acid contains oxides of
nitrogen which give a black colour with ferrous sulphate) carefully
into the liquid so as to form a heavy layer below. At the junction of
the liquids a black ring (purple if only traces of nitrate are present) is
formed. On shaking, the black colour disappears, bubbles of nitric
oxide are evolved, and a yellow solution of ferric sulphate remains.

Other tests for nitric acid are : (i) the red colour produced with a
solution of the alkaloid brucine in concentrated sulphuric acid ; (ii) the
deep blue colour with a solution of diphenylamine in concentrated
sulphuric acid ; (iii) the evolution of red fumes of oxides of nitrogen
when heated with concentrated sulphuric acid and copper turnings.

Dilute nitric acid is used as an oxidising agent in organic chemistry.
Thus, if toluene, C6H5-CH3, is boiled with the dilute acid, it is
oxidised to benzoic acid, C6H5*COOH.

Concentrated nitric acid, in the absence of water (e.g., in presence
of concentrated sulphuric acid), acts on many organic compounds
so as to replace one or more atoms of hydrogen by the nitro-group,
N02. This action is called nitration, and in it the acid behaves as
though it had the structural formula N02-OH.

Thus, benzene is converted into nitrobenzene : C6H6 + NO2-OH =
C6H5NO2 + H2O ; phenol yields, on prolonged nitration, trinitro-
phenol, or picric acid :

C6H6-OH + 3NO2-OH = C6H2(NO2)3-OH + 3H2O ;

570 INORGANIC CHEMISTRY CHAP.

toluene forms trinitrotoluene (T.N.T), C6H2(NO2)3-CH3, which, like
picric acid, is a powerful explosive.

EXPT. 229. — Shake a few drops of benzene with a mixture of con-
centrated nitric and sulphuric acids in a test-tube. Pour into water : a
yellow oil (nitrobenzene), smelling of bitter almonds, separates.

Glycerin and cotton (cellulose) do not form nitre-compounds
in the true sense, but salts of nitric acid with organic radicals, i.e.,
esters of nitric acid. Nitroglycerin is really glyceryl trinitrate :
C8H6(OH)8 + 3HN03 = C3H5(N03)3 + 3H20 ; nitrocelluloses, con-
taining from four to six N03 groups according to the concentration
of the acid, constitute collodion and gun-cotton ; the former is
soluble in a mixture of alcohol and ether : C12H20010 -J- 4HN03 =
C12H1606(N03)4 + 4H20.

The action of nitric acid on metals. — All metals, except platinum,
rhodium, iridium, and gold, are attacked by dilute or concentrated
nitric acid. Tin, antimony, tungsten, molybdenum, and arsenic
are converted into the oxides ; the rest form nitrates. During the
reaction a portion of the acid is reduced, with the formation of the
oxides N02, N2O3, NO, and N2O, free nitrogen, hydroxylamine,
and ammonia. The products depend on the metal, the tempera-
ture, the concentration of the acid, and the presence of the products
of reaction in the solution. Hydrogen is evolved only by the metal
magnesium, acting on cold dilute nitric acid : Mg -f 2HNO3 —
Mg(N03)2+JH2.

H. E. Armstrong and Ackworth (1877) suggested that the primary
reaction in all cases is the liberation of nascent hydrogen :
I. M -f HN03 — MNO3 -h H. On the ionic theory, this corresponds
with the reaction : M + H' — M' -f H. This nascent hydrogen,
however, is in contact with nitric acid, which is easily reduced, and
further reactions occur :

II. Secondary reactions, which probably proceed in definite stages :

(a) HNO3 + H2 = HNO2 (nitrous acid) + H2O.

(b) 2HNO3 + 4H2 = H2N2O2 (hyponitrous acid) + 4H2O.

(d) HNO3 + 4H2 = NH3 (ammonia) -f 3H2O ;

III. Tertiary reactions, in which the secondary products interact :

(1) by decomposition, giving simpler compounds :
(a) 3HNO2 = HNO3 + 2NO (nitric oxide) + H2O.
(6) 2HNO2 = N2O3 (nitrous anhydride) -f H2O.
(c) H2N2O2 = N2O (nitrous oxide) + H2O ;

(2) by double decomposition :

(a) HNO2 + NH3 = N2 (nitrogen) + 2H2O.

(b) HNO2 + NH3O = N2O + 2H2O.

xxix THE OXIDES AND OXY-ACIDS OF NITROGEN 571

The action of nitric acid on copper, on Armstrong's theory, would be
represented as follows :

I. 3Cu + 6HNO3 = 3Cu(NO3)2 + 3H2.
II. 3H2 + 3HNO3 = 3HNO2 + 3H2O.
III. 3HNO2 = HNO3 + 2NO + H2O.
/. by addition : 3Cu + 8HNO3 = 3Cu(NO3)2 + 2NO + 4H2O.

The reaction with zinc, which gives nitrous oxide, can be represented
as follows :

I. Zn + 2HNO3 = Zn(NO3)2 + H2.
II. (a) HN03 + 3H2 = NH3O + 2H2O.

(b) HNOS + H2 = HNO2 + H2O.
III. NH30 + HN02 = N20 + 2H20.

To obtain 4H2 we require 4Zn + 8HNO3, and 2HNO3 are reduced ;
hence : 4Zn + 10HNO3 = 4Zn(NO3)2 + N2O + 5H2O.

According to Divers, some metals give nitric oxide, but no hydroxyl-
amine or ammonia : e.g., Ag, Cu, Bi, Hg ; whilst other metals give
NH3, or NH3O, and N2O : e.g., Fe, Al, K, Zn, Sn, Cd, Mg (also gives H2).
The product, however, depends on the concentration and temperature
of the acid : thus concentrated nitric acid gives mainly nitrogen dioxide
with copper : Cu + 4HNO3 = Cu(NO3)2 + 2NO2 + 2H2O ; and also
on the accumulation of the salt in the solution, since by the prolonged
action of dilute nitric acid on copper, nitrogen is evolved.

Veley (1891) showed that pure nitric acid, in the absence of nitrous
acid, scarcely acts on copper, silver, bismuth, or mercury. Other
metals react in the absence of nitrous acid, but more slowly than
when it is present. Since nitrous acid is formed in the reaction,
the speed of the latter increases as it proceeds.

EXPT. 230. — Take three pieces of clean copper foil and immerse them
in three glasses containing : (a) 50 c.c. of 50 per cent, nitric acid ;
(b) 50 c.c. of this nitric acid -f- 5 c.c. of hydrogen peroxide
(20 vols.) ; (c) 50 c.c. of nitric acid -j- 1 gm. of urea. The foil in (a)
is at once violently attacked ; those in (b) and (c) remain for a time
without change. The hydrogen peroxide oxidises nitrous acid :
HNO2 -f H2O2 = HNO3 + H2O ; whilst urea decomposes it :
CO(NH2)2 + 2HN02 = C02 + 2N2 + 3H2O.

According to Veley, nitric oxide is a primary product, formed
from traces of nitrous acid ; a green solution of copper nitrite is
formed, which is then decomposed by nitric acid to reproduce
nitrous acid, most of which decomposes :

I. Cu + 4HN02 = Cu(N02)2 + 2H20 -f 2NO.
II. (a) Cu(N02)2 -j- 2HN03 = Cu(NO8)2 + 2HN02 ;
(b) 3HNOa =± HN03 + H20 + 2NO.

572

INORGANIC CHEMISTRY

CHAP.

The manufacture of nitric acid.— Nitric acid is made on the large
scale by three processes :

(1) The distillation of sodium nitrate with concentrated sulphuric

acid : NaN03 + H2S04 = NaHS04 + HN03 ; the retort
process.

(2) The direct combination of nitrogen and oxygen of atmospheric

air at the high temperature of the electric arc, and the
subsequent oxidation of the nitric oxide to nitric acid in
presence of excess of air and water :

4NO 4- 302 4- 2H2O = 4HN03 ; the arc process.
(3) The oxidation of ammonia, by passing a mixture of ammonia
gas and air over a catalyst, such as heated platinum, and
the subsequent treatment of the nitric oxide as in (2) :
4NH3 4- 502 == 4NO 4 6H20
4NO 4- 302 4- 2H2O = 4HN03.

Retort processes. — One or two tons of previously dried sodium

nitrate (Chile nitre)
are heated with rather
more than this weight
of concentrated sulphuric
acid (93 per cent. H2S04)
in a large cast-iron pot,
made in two or three
pieces clamped together
with asbestos packing and
built in a brickwork fur-
nace so as to allow of
very uniform heating
(Fig. 296). Nitric acid
vapour does not attack
iron, which is corroded
by the liquid acid. At
the top of the retort is a
manhole for introducing
the charge, and an outlet
for the acid vapour,
which is prolonged in a
stoneware, silicon-iron, or
FIG. 296.— Nitric Acid Retort. vitrified silica pipe, in

which there is a " lan-
tern," consisting of a stoneware box with glass windows, or a
short length of glass pipe, for observation. Twenty parts of coal

XXIX

THE OXIDES AND OXY-ACIDS OF NITROGEN

573

per 100 of nitre are required, and the distillation occupies about
fifteen hours.

The acid is condensed in some type of cooler : various forms are
used, consisting of vitrified silica spirals cooled in water, stoneware
U -tubes or horizontal glass tubes cooled partly by air and, water,
or S-shaped tubes of silicon iron. Large stoneware WouhVs
bottles are also used. One type of condenser (Fig. 297) consists
of earthenware tubes cooled by a shower of water. The red fumes
of oxides of nitrogen also produced are condensed by water in a
stoneware tower at the end, packed with hollow stoneware balls :
4N02 + 302 + 2H.O = 4HN03.

In the Valentiner process (1891) the whole apparatus is air-tight,

Retort

FIG. 297.— Guttmann Nitric Acid Plant.

and a vacuum is maintained at the end by an air-pump. The
distillation under reduced pressure (25 mm.) takes place at a lower
temperature (100°), so that there is less decomposition, and the
reaction also occurs more rapidly than in the ordinary process.

The liquid residue in the retort is run out from the lower pipe,
and allowed to solidify ; it is a mixture or compound of NaHS04
and Na2SO4, with a little NaN03, and is called nitre cake.

The arc process. — The union of atmospheric nitrogen and oxygen
at the high temperature of the electric arc was demonstrated by
Crookes ; a small experimental plant was worked at Manchester
in 1900. The foundation of the present industry, however, was
laid by the Norwegians, Birkeland and Eyde, in 1902. As at
present carried out in Norway, at Notodden and Riukan, the process
utilises 350,000 horse-power, all being derived from water-power.

574

INORGANIC CHEMISTRY

Spreading arc flame

Electrode

CHAP.

Air is drawn through a flat circular furnace (Fig. 298), in which an
electric arc, burning between water-cooled copper poles, is spread
out by an electromagnet into a disc of flame, the temperature of
which is about 3000°. In passing through this flame, combination
between the oxygen and nitrogen occurs : N2 + 02 •*— 2NO. At
3000° the equilibrium yield of NO is 5 per cent, by volume ; at
1500° it is only 04 per cent., since the reaction absorbs heat. The
gases leave the furnace at about 1000°, containing 1 per cent, of NO.
They pass through iron pipes lined with brick to the firebox of a

tubular boiler, where they are
cooled to 150°, with production of
steam, which is used to evaporate
•distributors solutions formed in the process.
The gas now passes through large
aluminium pipes exposed to the
air, where it cools to 50°.

When the gas has cooled below
600°, formation of nitrogen
dioxide begins : 2NO+02^±2N02;
this is a somewhat slow process,
and to give time for the reaction
to proceed the gases from the
air-coolers are passed through a
large empty iron oxidising tower.
From this the gas passes to the
first of three or four gigantic
absorption towers, built of granite
slabs, 65-80 ft. high and 18 ft.
diameter, packed with broken
quartz over which water is circulated. In these towers formation
of nitric acid occurs, involving the following reactions :

2N02 -f H20 = HN02 -f HN03.
-^203 + H20.

Magnet

FIG. 298.— Birkeland-Eyde Arc Furnace.

The N20g is evolved from the solution, then decomposes nearly
completely into NO and N02. The latter reacts over again, whilst
the NO is reoxidised by the excess of air present, forming N02,
which also enters into reaction. Nearly all the nitrous acid is
removed from the solution, and 30 per cent, nitric acid runs from
the first tower, the acid having been pumped from the final tower
through all the towers in succession.

The dilute nitric acid is either neutralised with limestone, to form
calcium nitrate, which is evaporated and exported as a fertiliser
(" Norge saltpeter ") ; or is concentrated by distilling it with con-
centrated sulphuric acid.

XXIX

THE OXIDES AND OXY-ACIDS OF NITROGEN

575

When the gases become very dilute, the oxidation of NO is very
slow, so that a mixture of NO and N02 passes from the last absorp-
tion tower, about 85 per cent, of the oxides of nitrogen having by
this time been absorbed. This is passed into an iron tower packed
with quartz, down which a solution of sodium carbonate trickles.
This absorbs nearly all the residual oxides, with formation chiefly
of sodium nitrite :

NO -f N02 ±^ N203 ; N203 -f 2NaOH - 2NaN02 + H20.

The oxidation of ammonia. — In 1795, the Rev. A. Milner, Fellow of
Queens' College, Cambridge, found that ammonia, when passed over
heated manganese dioxide, is oxidised to red fumes which on dis-
solving in water form nitric acid. The French chemist Kuhlmann,
in 1839, found that ammonia can be oxidised by passing it, mixed
with air, over heated platinum : 4NH3 -f 502 = 4NO + 6H20.
The colourless gas, on cooling, becomes red from further oxidation
of the nitric oxide : 2NO -f 02 = 2N02. It may be absorbed in
water, with formation of nitric acid, as described under the arc
process.

EXPT. 231. — Pass a current of air through ammonia in a wash-
bottle, and lead the mixed gas over a small roll of
platinum foil heated to dull redness in a hard
glass tube. Notice the formation of red fumes
in the globe attached to the tube (Fig. 270).

The best results are obtained when the gases
are passed very rapidly through the catalyst ;
with a slow current of gas the NO is broken up
again, or reacts with the ammonia : 4NH3 -|-
6NO = 5N2 -f 6H2O. In the latest type of
apparatus (Fig. 299) a mixture of 1 vol. of purified
ammonia gas and 7-5 vols. of air, filtered from
dust, is passed through two or three pieces of
fine platinum gauze stretched across a rectan-
gular aluminium box, and heated electrically.
The gases may also be heated to about 500°
before passing to the apparatus, and the reaction
then proceeds automatically. More than 90 per
cent, of the ammonia is oxidised to NO, and
the production is very large, since 1 sq. ft. of
combined catalyst gauze will effect the produc-
tion of 1-7 tons of UNO 3 per twenty -four
hours. The gases are cooled, and treated in
towers as in the arc process.

FIG. 299. — Ammonia
Oxidation Converter.

570 INORGANIC CHEMISTRY CHAP.

If the cooled gas is passed through milk of lime, calcium nitrate
is produced : the first reaction is

2Ca(OH)2 + 2N2O4 = Ca(NO3)2 -f Ca(NO2)2 + 2H20.

When all the lime is neutralised, nitric acid is formed in the solution
by reactions previously explained. This decomposes the nitrite,
with evolution of oxides of nitrogen, which are fully oxidised to
N02 by air and passed into another absorber of milk of lime :
Ca(N02)2 + 2HN03 = Ca(N03)2 + NO + NO2 + H.,< >.

If ammonia gas mixed with air is blown into the cooled and
fully oxidised gas from the oxidation apparatus, solid ammonium
nitrate is deposited as a powder : 4N02 + O2 + 2H2O + 4NH3 =-
4NH4N03.

The nitrogen cycle. — Nitric acid is formed by electrical dischai"<
in the atmosphere, and is washed down by rain. It is estimated
that no less than 250,000 tons of nitric acid are so produced in
twenty-four hours; only a small amount of this falls on fertile
soil, and is utilised by plants. Besides the nitric acid produced
by electrical discharges, which is absorbed from the soil in the form
of nitrates by plants, it is found that leguminous plants can grow
and form organic nitrogen compounds (proteins) in air and soil
free from ammonia or nitrates. These plants take up atmospheric
nitrogen, which is converted into organic nitrogen by the agency
of micro-organisms called bacteroids, which occur in nodules on the
root-hairs. Algae, fungi, and mosses are also capable of utilising
elementary nitrogen. The organic nitrogen compounds elaborated
by plants serve as food for herbivorous animals, and the proteins
of the latter are utilised in turn by carnivora.

When the bodies of animals and plants decay, or are subjected
to destructive distillation, ammonia is produced. In the soil this
is oxidised by nitrifying bacteria to nitrites, and nitrates, the latter
again serving for the nourishment of plants. A portion of the
nitrogen, however, is again set free by the action of denitrifying
bacteria.

The nitrogen cycle in Nature may be represented diagrammatically
as shown on page 577.

The so-called nitrogen problem arises from : (1) the former very
large dependence of civilised countries on the single source of supply
of nitrates in Chile; (2) the certain exhaustion of these deposits
in the near future. All civilised countries have taken steps to
render themselves more or less independent of external sources of
supply of nitrates : Germany is completely independent in this
respect.

TIIIO OXIhlOS AND OXY-ACIDS OK NITK()(JIO\?
NITROGEN CYCLE.

legUminOUS plants -\- buetemi,
101,
ATMOSIMIKKK:

NITUOUKN

r,77

r,u-( uirtcnurgoH > — •>. \,

V^X NITROGEN

1 )«-mti-i!\ ing
hue tori u

decay of

nitrifying
bacteria

plants and
a u i mala ;

desl ruclixe

distillation

nitrifying

bacteria

Nitric anhydride, nitrogen pentoxide, N205. The anhydride <l

nitric :icid was obtained by Deville (IS 19) by (he action of ehlorine
on silver nitrate : 4AgNO, I ^ 'L 4AgCl | L> N ,< )f) f O2. It is

more conveniently prepared by dehydrating concentrated nitric
acid by phosphorus pentoxide (Weber) : 2HN08= N2()6 -f H,<>.

T.I concentrated pun- nitric acid in a stoppered retort, cooled in a
fivey.iiitf mixture, pui-e phosphorus piM»toxid(^ is a.ld.-d in slight excess
in small (jiiantilies ut u tini(\ The mixture is allowed to stand and
di.-.hlled at a I emperat ure of :{0". Th(^ distillate in the cooled re-
ceiver consists of two layers : the upper, orange-red, layer solidifies
on moling in u free/in^ mixture to colourless crystals of N"aOr(. if the
di.-iillaiion is carried out in a current of ozonised oxygon, and the
gases are passed through a phosphorus pentoxide tube, perfectly pure
<-r\ l.i I <.l \ .< ) are obtaim-d by cooling in solid ( '( ).,, and et her.

Crystalline nitro^m pentoxide is also formed by passing ozonised
o\\<jen through cooled liquid nitrogen tetroxide : N204-{-08 =
Nji, | Oa,

Nitrogen pentoxide forms white, very hygroscopic, crystals,
\\hi( h are stable below 0 , but decompose slowly at the ordinary

tetnperaliire, even in sen led tubes, becoming yellow I 2N2(>5 --=

N.O4-|-02. The crystals melt with decomposition at 29-5 , and
form a dirk brown liipiid, which decomposes into red fumes of N()2
and oxygen at K» If suddenly heated, the ei'ystals explode ; they

dissolve with a hissing noise in water, forming nitric acid :
N .< > , HaO = 2HNOo. Phosphorus and potassium burn in the
liquid pentoxide if slightly warmed; charcoal does not decompose

it even on boiling, but burns brilliantly if previously ignited. Sul-

phur forms white vapours, condensing to crystals of mtrosulphonic
anhydride, SJ )5(N ().,).,. A crystalline compound, N205,2HN03,

578

INORGANIC CHEMISTRY

CHAP.

m.-pt. 5°, is formed on cooling a solution of the anhydride in con-
centrated nitric acid.

The formula of the anhydride may be written

° ®

nitric acid is HO — N^ , and the compound of the two may be :

N02— 0— NOV —OH

0
N02— 0— NO/— OH.

In considering the remaining oxides of nitrogen, it will be most
convenient to begin with nitric oxide. NO, which is involved in the
preparation of some of the other oxides.

Nitric oxide, NO. — Although nitric oxide appears to have been
obtained by Mayow, Hales, and Cavendish, it was first recognised
as a distinct gas by Priestley (1772), who prepared it by the action
of copper or mercury on dilute nitric acid, and called it nitrous air :
3Cu + 8HNO3 = 3Cu(N03)2 -f 2NO + 4H2O.

EXPT. 232. — Copper turnings are placed in a flask (Fig. 300) and a
mixture of equal volumes of concentrated nitric acid and water (sp. gr.

1-2) is poured on. At first,
the flask becomes filled with
red fumes, due to the action
of the nitric oxide on the
atmospheric oxygen : 2NO +
O2 = 2NO2. When these are
driven out, the gas becomes
nearly colourless, but always
has a slight yellowish tinge,
since a little NO2 is produced
by the action of the metal on
the acid. This colour is
removed when the gas is
passed through water, and
the jars fill with a colourless
gas only slightly soluble in water. The gas so prepared, especially in
the later stages of the reaction, contains a variable amount of nitrogen.
It may be purified by passing into a cold saturated solution of ferrous
sulphate. A nearly black liquid is formed, containing FeSO4-NO, which
on gentle heating evolves nearly pure nitric oxide. The gas so
purified still contains 1/500 of its volume not absorbed by fresh
ferrous sulphate.

PIG. 300. — Preparation of Nitric Oxide.

XXIX

THE OXIDES AND OXY-ACIDS OF NITROGEN

579

Nearly pure nitric oxide may be obtained by heating a mixture
of potassium nitrate, ferrous sulphate, and dilute sulphuric acid.
A dark brown solution of NO in ferrous sulphate is first formed,
which breaks up on heating :

KN0 H2S04 = KHS04 + HN0

3FeS04 + 2HN03 + 3H2SO4 = 3Fe2(S04)3 -f 2NO

4H20.

If a solution of iron in concentrated hydrochloric acid is mixed
with an equal volume of the acid, and the solution heated with
sodium nitrate, nearly pure nitric oxide is evolved :

3FeCl2 + NaNO3 -f 4HC1 = 3FeCl3 -f NaCl + 2H2O + NO.

Perfectly pure nitric oxide is obtained (W. Crum, 1840) by shaking
mercury in a flask with concentrated sulphuric acid to which sodium
nitrate has been added ; the gas is purified by
passing over solid potash. 2HN03 -j- 6Hg -f-
3H2S04 = 2NO + 3Hg2S04 + 4H20.

This reaction is used in the estimation of
nitrites or nitrates, or of oxides of nitrogen in
commercial sulphuric acid. The substance is
dissolved in the least amount of water and
passed into the Lunge nitrometer (Fig. 301), which
consists of a graduated tube, A, with a stopcock,
B, communicating with a small cup, C, and an
outlet tube, D, the whole being filled with mer-
cury and provided with a levelling tube, E.
Concentrated sulphuric acid is then introduced, and
the mixture shaken violently with the mercury.
The volume of nitric oxide is read off.

Pure nitric oxide is evolved by dropping a
solution of potassium nitrite and potassium
ferrocyanide into dilute acetic acid :
K4FeC6N6+KN02+2CH3-COOH=K3FeC6N6+
2CH3-COOK + H2O -f NO. The gas should
be collected over mercury, as it acts slightly
on water, evolving traces of nitrous oxide.

Properties of nitric oxide. — Nitric oxide is
a colourless gas, slightly heavier than air
1 -3402 gm. /lit.), and sparingly soluble in water :

'Temp. ...... 0° 15° 30° 60°

Vols. of NO in 1 vol. of water 0 -074 0 -051 0 -040 0 -029

It is difficult to liquefy : the liquid boils at — 153°, and freezes at
- 167° to a white solid. The critical temperature is — 93-5°, and
the critical pressure 71-2 atm.

Nitric oxide is freely soluble in cold ferrous sulphate solution,
forming a black liquid, as was observed by Priestley. The maximum

p p 2

FIG. 301.— Lunge's
Nitrometer.

(normal density

580 INORGANIC CHEMISTRY CHAP.

absorption corresponds with FeSO^-NO, but the reaction is re-
versible, the absorption depending on the temperature, the concen-
tration of the ferrous salt (other ferrous salts, e.g., FeCl2, also
absorb NO, in different amounts), the pressure, and the presence of
other salts : FeSO 4 + NO ^= FeSO 4-NO. The gas is readily evolved
on heating. Manchot regards the compound as [Fe(NO)]S04^±
FeNO" -f SO/. The cation carries the nitric oxide with it on
electrolysis.

Nitric oxide is also absorbed by an acidified solution of potass-
ium permanganate : 6KMn04-f 10NO + 9H2S04 = 3K2SO4 +
6MnS04 4- 10HN08 + 4H2O.

It is not absorbed by alkalies, but dissolves in a mixture of caustic
soda and sodium sulphite, forming sodium nitrosohydroxylamine
sulphonate Na2(NO)2S03 or ON'N(ONa) S03Na.

Nitric oxide combines with free oxygen to form red fumes of
nitrogen dioxide : 2NO + 02 = 2N02. The reaction is not complete
unless a short time of contact is allowed : this is less than a second
with the pure gases, but may amount to several minutes with very
dilute mixtures of nitric oxide and air. In contact with water, the
red fumes dissolve, forming a mixture of nitrous and nitric acids :
2NO2 + H2O = HN02 + HN03. If the nitric oxide and oxygen
are dried with phosphorus pentoxide they do not combine.

Some combustible substances burn in nitric oxide, but the material
must first be ignited in air, and then introduced, freely burning, into
the nitric oxide. The latter is the most stable oxide of nitrogen :
it begins to- decompose into nitrogen and oxygen appreciably only
above 1000°, and unless this temperature is attained combustion does
not proceed. The substances burn only in the oxygen liberated by
the thermal decomposition of the gas. A lighted taper, burning
sulphur, and charcoal are extinguished in the gas. Feebly burning
phosphorus is also extinguished in the gas. but if burning brightly the
combustion continues brilliantly, red fumes being produced as well
as white clouds of phosphorus pentoxide : 2NO = N2 + 02 ;
?4 + 5O2 = 2P205 ; 2NO -f O2 = 2N02. A mixture of carbon
disulphide vapour and nitric oxide burns with a brilliant blue flame
(p. 730).

A mixture of hydrogen and nitric oxide when passed over heatec
platinum black is reduced to ammonia : 2NO -f- 5H2 = 2NH3
2H2O. Higher oxides of nitrogen, and nitric acid vapour,
similarly reduced.

Nitric oxide is absorbed by nitric acid ; with concentrated acic
a yellow solution of N02 is obtained. With more dilute acid a blu<
(N2O3) or green (N02 + N203) solution is formed, the blue solutioi
being obtained with the most dilute acid. Beyond a certain dilutic
the acid absorbs very little of the gas.

The composition of nitric oxide may be determined by heating

XXIX

THE OXIDES AND OXY-ACIDS OF NITROGEN

581

spiral of iron wire, by an electric current, in a measured volume of
gas. The apparatus shown in Fig. 302 may be used. The oxygen is
removed by the iron and, after cooling, half the volume of nitrogen
is left. The density of the gas is 15 (H = 1), hence the molecular
weight is 30. This contains half its volume, or 14 parts, of nitrogen,
and 30 - 14 = 16 parts of oxygen, i.e., 1 atom of each element, so
that the formula is NO. Nitric oxide
does not explode with hydrogen unless
previously mixed with an equal volume
of nitrous oxide.

The analysis of nitric oxide by heating
finely-divided nickel in the gas was care-
fully executed by R. W. Gray (1905).
The ratio was : — N : O = 14-0085 : 16.
The density of the gas was also found
to be 1-3402, so that, after a correction
for compressibility, the molecular weight
(O = 16) = 30-009 ; or N = 30-009 — 16 =
14-009.

The apparatus used in the analysis
of NO is shown in Fig. 303. The gas
was contained in the bulb A, which was weighed, first empty and
then full of gas. The platinum boat, H, heated by a platinum spiral,
contained the nickel. The bulb M contained charcoal. After the

decomposition was
complete, the bulb
M was put in com-
munication with A
and immersed in
liquid air. The nitro-
gen condensed on the
charcoal, and was
weighed. The weight
of A now gave the
weight of oxygen
which had combined
with the nickel :
2NO + 2Ni = N2 +
2NiO.

FIG. 302. — Composition of
Nitric Oxide.

FIG. 303.— Gray's Apparatus for determining the
Composition of Nitric Oxide.

Nitrous oxide,N20.

-Priestley (1772)
noticed that if nitrous air (NO) is allowed to stand in contact
with moist iron filings, or liver of sulphur, it contracts, like
common air, but the residual gas differs completely from that left
by common air (N2) in supporting combustion vigorously.
Priestley called the -gas diminished nitrous air. The gas was
carefully examined by Davy in 1799. He first prepared it in the

582 INORGANIC CHEMISTRY CHAP.

pure state, by heating ammonium nitrate, determined its com-
position, and examined its physiological action. He calle$ it
nitrous oxide. Its use as an anaesthetic and its peculiar effects
(" laughing gas ") are well known.

Nitrous oxide is produced by the reduction of moist nitric oxide
by sulphur dioxide or sulphites, or of nitric acid by metals or
stannous chloride under special conditions :

4Zn + lOHNOo (dilute) - 4Zn(N03)2 + 5H20 -f N20.

2HNO3 + 4SnCl2 + 8HC1 = 4SnCl4 + 5H2O +N2O.

2NO + S02 + H2O = H2S04 + N20.

A purer gas is more conveniently obtained by the decomposition
of ammonium nitrate by heat (Davy) : NH4N03 = N2O -j- H.,0.
Very pure nitrous oxide is obtained by heating a solution of equi-
molecular amounts of hydroxylamine hydrochloride and sodium
nitrite : NH3O + HNO2 = N2O + 2Ha(X

EXPT. 233. — Heat about 50 gm. of pure ammonium nitrate, pre-
viously dried at 105°, in a glass retort over wire gauze. The salt melts
at 170° (when quite dry ; usually at 165°), and begins to decompose
below 200°. The reaction is exothermic : NH4NO3 = N2O + H2O +
25 kg. cal., and if the salt is heated above 250° it is liable to explode :
before this occurs, nitric oxide, nitrogen, and ammonia are evolved.

The gas is purified from higher oxides of nitrogen by passing through
potassium permanganate solution, from chlorine (derived from ammo-
nium chloridQ in the ammonium nitrate) and nitric acid vapour by caustic
soda, and from ammonia by concentrated sulphuric acid, and is collected
over hot water or mercury.

The nitrate may be mixed with three parts of sand and heated to
260-285° ; a mixture of 2 molecular proportions of NaNO3 with 1 of
(NH4)2SO4 on heating to 240° evolves a regular stream of nitrous oxide.

Nitrous oxide is prepared for use as a mild anaesthetic ; it is
liquefied by compression in steel cylinders. The gas should be
carefully purified from chlorine and nitric oxide, as described
above.

Properties of nitrous oxide. — Nitrous oxide is a colourless gas,
normal density 1 -9777 gm. /lit., with a faint sweetish odour and taste.
It is appreciably soluble in water :

Temp 0° 5° 10° 15° 20° 24C

Vols. of N2O in

1vol. of water 1-3052 1-0954 0-9196 0-7778 0-6700 0-5962

The solution has no action on litmus, so that the gas does not
behave as the true anhydride of hyponitrous acid : H2O -f- N2O =
H2N2O2. When cooled to — 90°, or exposed to pressure (30
atm. at 0° ; 40 atm. at 15°), it forms a colourless mobile liquid,

xxix THE OXIDES AND OXY-ACIDS OF NITROGEN 583

b.-pt. —88-7°; the critical temperature is 354°, the critical
pressure 75-0 atm. The liquid is lighter than water (sp. gr. 0-908) ;
when cooled to —115°, or when rapidly evaporated (not sponta-
neously on reducing the pressure, as in the case of liquid carbon
dioxide), it forms a snow-like mass, with some transparent crystals
of the solid, m.-pt. -102-3°.

Nitrous oxide supports combustion more vigorously than air,
since it yields on decomposition a gas containing one-third its
volume of oxygen, as compared with one-fifth in air : 2N2O =
2N2 + 02. Nitric oxide gives a gas containing half its volume of
oxygen, but does not support combustion so well as air or nitrous
oxide. This arises from the circumstance that nitrous oxide is
much more easily decomposed by heat than nitric oxide : the
latter is stable to about 1000°. Decomposition of nitrous oxide
begins at 520°, and is complete at 900°. The gas is also decomposed
by sparks, but some nitric oxide is also formed, presumably by
recombination of nitrogen and oxygen. All combustions in nitrous
oxide are really combustions in the oxygen set free on heating the gas.

EXPT. 234. — A taper burns in the gas with a brilliant flame, and a
glowing chip is rekindled as in oxygen. Nitrous oxide, however, is
distinguished from oxygen by its smell, its greater solubility in water,
and the fact that it does not produce red fumes with nitric oxide.

EXPT. 235. — Brightly burning phosphorus burns in the gas with a
brilliant flame, producing clouds of pentoxide, and a little red fume of
nitrogen dioxide. ( How is the latter formed ?) Feebly burning sulphur
is extinguished, but if brightly burning, the sulphur continues to burn
vigorously with a double flame. The outer, large, flickering, yellow
flame corresponds with the reaction 2N2O = 2N2 + O2, and the inner,
bright blue flame to the reaction S -f O2 = SO2. Sodium and potass-
ium burn in the gas to form peroxides, and iron wire burns as in oxygen.

Nitrous oxide is an endothermic compound, and is decomposed into
its elements by the shock of exploding fulminating mercury. If
mixed with detonating gas (2H2 + 02), nitrous oxide is also
completely decomposed on explosion, and this may be used to
determine the composition of the gas.

Two vols. of nitrous oxide when mixed with electrolytic gas and
exploded leave three volumes of gas (all the electrolytic gas is condensed
to liquid water). On treatment with pyrogallol and caustic potash, 1
vol. of oxygen is absorbed, and 2 vols. of nitrogen are left. Davy
determined the composition of nitrous and nitric oxides by heating
potassium in a measured volume of the gas confined in a bent tube over
mercury. After cooling, an equal volume of nitrogen remained. The
gas may also be decomposed by a heated spiral of iron wire, as in the

584 INORGANIC CHEMISTRY CHAP.

case of nitric oxide : in this way Jaquerod and Bogdan (1904) found
that 1 vol. of N2O gave 1 -00686 vols. of N2.

These experiments show that nitrous oxide contains its own
volume of nitrogen. The relative density of the gas (H = 1) is 22,
hence the molecular weight is 44. But this contains a molecular
weight (i.e., an equal volume) of nitrogen, N2, of weight 28, and there-
fore 44 — 28 = 16 parts, or one atom of oxygen. The formula is
therefore N2O.

The formula may also be established by exploding the gas with
hydrogen in a eudiometer. If 20 c.c. of nitrous oxide are mixed with
20 c.c. of hydrogen and exploded, 20 c.c. of nitrogen are left. The
hydrogen must have combined with 10 c.c. of oxygen to form liquid
water, so that 2 vols. of nitrogen are combined in 2 vols. of nitrous
oxide with 1 vol. of oxygen, and the formula is N20.

Nitric oxide does not explode with hydrogen, but if mixed with an
equal volume of nitrous oxide both gases explode when sparked with
hydrogen.

In an experiment a mixture of 20 c.c. of nitrous oxide, 20 c.c. of nitric
oxide, and 40 c.c. of hydrogen was exploded. Thirty c.c. of nitrogen
remained. Of this, 2£) c.c. must be derived from the nitrous oxide :

N20 + H2 = N2 + H20;
20 c.c. 20 c.c. 20 c.c.

hence the 20 c.c. of nitric oxide gave 30 — 20 = 10 c.c. of nitrogen.
Again, 20 c.c. of hydrogen are used up by the nitrous oxide, so that
40 — 20 = 20 c.c. of hydrogen have combined with the oxygen in the
20 c.c. of nitric oxide, which must therefore have been 10 c.c. Thus,
20 c.c. of nitric oxide contain 10 c.c. of nitrogen and 10 c.c. of oxygen ;
this corresponds with the formula NO :

02 + N2 = 2NO
10 c.c. 10 c.c. 20 c.c.
N20 + H2 = N2 + H20
20 c.c. 20 c.c. 20 c.c.

NO + H2 = |N2 + H20
20 c.c. 20 c.c. 10 c.c.

40 c.c. 40 c.c. 30 c.c.

Nitrous acid and nitrites. — Scheele (1772) observed that the residue
left after heating nitre effervesced with acids and gave red fumes,
hence he concluded that it was a salt of a new acid. The residue
is potassium nitrite : 2KN03 = 2KNO2 -f O2. The reduction is
effected at a lower temperature by fusing potassium or sodium
nitrate with lead or copper, lixiviating with water, filtering from the
metallic oxide, and evaporating : NaN03 -f- Pb = NaNO2 -f- PbO.
A little caustic soda is formed, which dissolves lead oxide. This is

xxix THE OXIDES AND OXY-ACIDS OF NITROGEN 585

precipitated by a stream of carbon dioxide, or by carefully neutralis-
ing the liquid with nitric acid. The crystals of sodium nitrite are
dried in a centrifugal machine, then in an oven at 50°. They have
a yellowish colour, and always contain a certain amount of nitrate.
Potassium nitrite may be obtained similarly, but does not crystal-
lise well, hence it is precipitated from the solution by alcohol, or
fused and cast into sticks.

Purer nitrites are formed by passing the red fumes evolved on
heating nitric acid with arsenious oxide (p. 587), and consisting of a
mixture of equimolecular amounts of nitric oxide and nitrogen
dioxide, NO -f- N02, probably in equilibrium with a small quantity of
nitrous anhydride, N2O3, into a solution of caustic potash or soda
(sp. gr. 1 *38), or their carbonates, out of contact with air : 2KOH -f-
(NO -f- N02) = = 2KN02 + H20; Na2CO3 + (NO + N02) =
2NaN02 -f- C02. Pure potassium nitrite is obtained by decomposing
amyl nitrite with alcoholic potash : C5H11NO2 -j- KOH =
C5Hn-OH (amyl alcohol) -f KN02.

Both potassium and sodium nitrites are slightly yellow and their
concentrated solutions are markedly yellow. The solutions are
alkaline, owing to hydrolysis, since nitrous acid is a weak acid :
N02' -f- H20 ^ HN02 + OH'. Sodium nitrite fuses at 213°, ; and
at 15°, 5 parts of NaN02 dissolve in 6 parts of water. Its crystals
are thin flat prisms, moderately deliquescent ; it may be purified
by recrystallisation (unlike KNO2). Potassium nitrite occurs in
minute short prisms, containing no water but exceedingly deli-
quescent, and soluble in one-third the weight of water.

Barium nitrite may be obtained as above, using baryta water, but is
more conveniently prepared by mixing hot, almost saturated, solutions
of sodium nitrite and barium chloride, filtering off the sodium chloride
in a hot-water funnel, and allowing the filtrate to crystallise : 2NaNO2 +
BaCl2 ~ ^ 2NaCl + Ba(NO2)2. The salt is recrystallised, and dried over
sulphuric acid, when it forms Ba(NO2)2,H2O.

Silver nitrite, AgNO2, is obtained as a white, sparingly soluble pre-
cipitate, when an alkali nitrite is added to silver nitrate solution. It
is purified by recrystallisation from hot water.

If dilute sulphuric, hydrochloric, or even acetic acid is added
to a solution of a nitrite, free nitrous acid, HN02, is first formed, but
is almost completely decomposed with effervescence, red fumes of
oxides of nitrogen being liberated. The solution has a pale blue
colour, which appears to be due, not to nitrous acid, but to nitrous
anhydride, N2O3 ; this has a deep blue colour in the liquid state.

The blue colour is also communicated to chloroform when shaken
with the aqueous solution, although the latter can never be quite
decolorised. The decomposition of the nitrous acid in fairly con-

586 INORGANIC CHEMISTRY CHAP.

centrated solutions probably occurs according to the equation :
2HN02 =± N2O3 + H2O ^± NO + N02 -f H20. In dilute solutions
it may decompose according to the equation : 3HNO2 ^± HN03 +
2NO -f- H20, although this may be regarded as the result of
the following reactions : (a) 4HN02 :=± 2NO + 2N02 -f 2H20 ;
(6) 2N02 + H20 ^± HNO2 + HN03. The amount of nitrous acid or
its anhydride left in aqueous solution never exceeds a few per cent.
A pure dilute solution of nitrous acid is obtained by precipitating a
solution of barium nitrite with dilute sulphuric acid ; it is pale blue
in colour, and slowly decomposes, especially on heating, or shaking,
with evolution of nitric oxide.

Nitrous acid and nitrites act as reducing agents : HNO2 -f- O =
HN03 ; thus they reduce permanganates and chromates. They
may be estimated in solution by running into excess of warm acidified
standard potassium permanganate (e.g., N /2), and titrating the
latter with standard oxalic acid : 2KMn04 + 5HNO2 + 3H2SO4 =
K2S04 -f 2MnSO4 + 5HN03 -f 3H20. They are also oxidised
by bromine water : HN02 + Br2 + H20 = HN03 + 2HBr.

Nitrous acid and nitrites may sometimes act as oxidising agents :
2HNO2 = 2NO + O -}- H2O. In presence of atmospheric oxygen
and water, NO will reproduce nitrous acid, so that a small amount of
nitrous acid may effect a considerable amount of oxidation by acting
as a carrier of oxygen. Thus, iodine is liberated from potassium
iodide : 2KI + 2HN02 = 2KOH + I2 + 2NO, indigo is bleached,
stannous chloride is oxidised to stannic chloride : SnCl2 4- 2HC1
+ 2HN02 = 'SnCl4 + 2NO -f 2H20, sulphur is precipitated from
sulphuretted hydrogen, and sulphur dioxide is oxidised to sulphuric
acid. The free acid can be titrated with caustic soda and alizarin
red.

The liberation of iodine from potassium iodide (blue colour with
starch) serves as a delicate test for nitrous acid (or a nitrite in acid
solution). Still more delicate tests are the brown colour with a
solution of metaphenylenediamine hydrochloride in hydrochloric
acid ; and the intense pink colour with a mixture of solutions of
sulphanilic acid and a-naphthylamine in acetic acid. These two
reactions may be used for the detection and estimation of nitrites
in water.

Ammonium nitrite is prepared in solution by the decomposition of
barium nitrite by ammonium sulphate. The solution deposits crystals
when evaporated in a vacuum at the ordinary temperature. The solid
may also be prepared by acting with the red fumes (NO + NO2) from
nitric acid and arsenious oxide on lumps of ammonium carbonate
(p. 801), extracting the nitrite with absolute alcohol, and precipitating
the solution with ether. The crystals are deliquescent, and explode
when heated to 70° : NH4N"O2 = N2 + H2O.

xxix THE OXIDES AND OXY-ACIDS OF NITROGEN 587

The constitution of nitrous acid.— If ethyl alcohol is distilled with
sodium nitrite and sulphuric acid, ethyl nitrite, a colourless mobile
liquid with a pleasant odour, b.-pt. 17°, is obtained. On boiling with
caustic soda this hydrolyses, with the formation of ethyl alcohol and
sodium nitrite. The ethyl group appears therefore to be attached to
oxygen, not to nitrogen : O : N-O-C2H5 -f- NaOH = O : N-O'Na +
C2H5-OH. When heated with tin and hydrochloric acid, ethyl nitrite
is reduced to ammonia (and hydroxylamine) and alcohol :
O : N-OEt = H2O + NH3 + EtOH.

2H3H H

If silver or sodium nitrite is heated with ethyl iodide in a sealed tube,
nitroethane, C2H5NO2, isomeric with ethyl nitrite, but boiling at 113-114°,
is obtained. This is not hydrolysed by caustic soda, but an atom
of hydrogen in the ethyl group, C2H5, is replaced by sodium : it there-
fore behaves as acidic hydrogen, indicating that the ethyl group is
attached to a negative group (p. 517), NO2. On reduction with nascent
hydrogen, the ethyl group remains attached to the nitrogen atom, and
ethylamine, C2H6-NH2, which can be obtained by the action of ethyl
iodide on ammonia, and therefore has the above formula, is obtained.
These reactions indicate that the second compound has the formula

" 26,

IxN" — C2H6 ; it is nitroethane, i-e-, ethane, C2H, with an atom of

hydrogen substituted by the nitre-group, NO2 (p. 569). The reduction
may then be formulated : C2H6-NO2 + 6H = C2H5-NH2 + 2H2O.

Since both compounds may be obtained from sodium nitrite, the
latter behaves as a tautomeric compound, and there are therefore two
tautomeric forms of nitrous acid :

This example shows that evidence of the constitution of inorganic
compounds which is based on the reactions of organic compounds
must be accepted with caution.

Nitrous anhydride, or nitrogen trioxide, N203. — Red vapours are
obtained by distilling diluted nitric acid with arsenious oxide or
starch : 2HN03 + As203 = As205 + H20 + N203. On cooling the
vapours in a freezing mixture, a dark blue volatile liquid is obtained.

EXPT. 236. — Heat 100 gm. of white arsenic with 80 c.c. of nitric
acid of sp. gr. 1 -35 (56 per cent. HNO3) in a large flask with a long tube
bent slightly backward, as shown, and connected by an ordinary cork
with a glass worm cooled with ice and salt (Fig. 304). A deep blue
liquid condenses, and is collected in a tube contained in ice and salt.

588

INORGANIC CHEMISTRY

Vapours of higher

The tube may be sealed off to preserve the liquid.
oxides of nitrogen are dangerously poisonous.

The red gas is absorbed completely by caustic soda, either solid
or in solution, with formation of pure nitrites, and by concentrated
sulphuric acid, with formation of nitrososulphuric acid. It there-
fore behaves as if it were nitrous anhydride, N203:

2NaOH + N203 = 2NaN02 + H2O.

2S02(OH)2 + NO-O-NO = 2S02(OH)-0-NO + H20.

On the other hand, the vapour density shows that the gas is a
mixture of equal volumes of nitric oxide and nitrogen dioxide, so that

the compound N203
is apparently com-
pletely dissociated
into NO and N02.
Hasenbach,by pass-
ing the vapour
from the blue liquid
through a red-hot
tube, and then
through a freezing
mixture, obtained a
deep blue liquid, the
vapour of which,
when passed over
red-hot copper, gave
36 per cent, of
nitrogen, whilst
N203 requires 36-8
per cent. It was
therefore considered that nitrous anhydride, although it exists in
the liquid state, is completely dissociated as a gas.

Nitrogen dioxide dissociates on heating above 600° : 2NO2 =
2NO + O2. If the hot gas is rapidly cooled, oxidation of half the NO
occurs rapidly, producing NO + NO2, and as the further oxidation
occurs slowly, the gas when passed into a cold tube condenses as N2O3.

Ramsay and Cundall (1885) collected gaseous nitrogen dioxide in a
tube over mercury, and introduced into it a thin bulb filled with nitric
oxide. When the latter was broken there was no change of volume,
whereas, according to the experimenters, there should have been a con-
traction if N2O3 is formed :
NO +NO2=N2O3 (contraction £) ; 2NO+N2O4 = 2N2O3 (contraction A).

Dixon and Peterkin (1899) pointed out that if there had been no com-
bination an expansion of nearly 10 c.c. should have occurred, due to
dissociation of N2O4 present in the dioxide owing to its dilution with the

FIG. 304. — Preparation of Nitrogen Trioxide from
Arsenious Oxide and Nitric Acid.

xxix THE OXIDES AND OXY-ACIDS OF NITROGEN 589

other gas: N2O4 z=± 2NO2. Since there was really a contraction of
about 0-3 c.c., there must have been some reaction leading to diminution
of volume, which they assumed to be formation of N2O3. With nitrogen
dioxide and an indifferent gas, or with NO above 50°, there was the
normal expansion of 10 c.c. The gas obtained by mixing 100 vols. of
NO and 100 vols. of nitrogen dioxide (NO2 and N2O4) at 27° they calcu-
lated should have the following composition :

N204. N02. NO. N203. Total.

Before mixing ... 68 32 100 0 200

After mixing 62 38 94 6 200

If the blue liquid is dried by prolonged exposure to phosphorus
pentoxide it may be volatilised without decomposition, and has a
vapour density corresponding with N406 ; in presence of the least
trace of moisture it decomposes : N2O3 ^± NO -f- NO2 ; a little
N203 still remains in equilibrium, and it is this which causes the
reactions described above. As the trace of N2O3 is removed by
absorbents, it is quickly reproduced, since the equilibrium is
disturbed.

Liquid nitrous anhydride is obtained by the action of nitric oxide
on solid nitrogen dioxide cooled in liquid air. It is not oxidised to
N02 by oxygen below — 100°, solidifies at — 103°, and (unless quite
dry) begins to decompose at —21°.

If nitric oxide is mixed with air or oxygen, and the gas immediately
brought in contact with absorbents, it behaves as N203 (see above).
If it is allowed to stand a few minutes, it behaves as nitrogen dioxide :

(i) 2NO + 02 = 2N02 ; (ii) NO + N02 ^ N2O3 ; (iii) 2NaOH -f
N203 = 2NaN02 -f- H20 : rapid absorption.

(i) 2NO -+ O2 = 2NO2 (completely) ; (ii) 2N02 + 2NaOH =
NaNO2 -f NaN03 -f- H20 : after standing.

These reactions have often been interpreted as if N2O3 were the first
product of the oxidation of NO by oxygen, and was then further oxidised
to NO2. There is no evidence that this is the case ; all the reactions may
be explained by the slowing down of the speed of oxidation of NO to
NO2 when half the oxidation has been effected.

EXPT. 237. — To 40 c.c. of NO in a graduated tube over mercury
containing 20 c.c. of concentrated potash solution add rapidly 50 c.c.
of air. Almost immediate absorption of the red fumes occurs, and
40 c.c. of nitrogen are left (4NO + O2 + 4N2 ^ 2N2O3 + 4N2). To 20
c.c. of nitric oxide contained in a second tube, without alkali, add 50
c.c. of air. After standing for two minutes add 20 c.c. of potash solution.
The red fumes are more slowly absorbed than in the first experiment, and
40 c.c. of nitrogen are left (2NO + O2 + 4N2 = 2NO2 + 4N2). (Gay-
Lussac, 1816).

590

INORGANIC CHEMISTRY

CHAP.

EXPT. 238. — By means of a T-tube admit a small amount of NO
from a gas-holder into a rapid stream of air passing into a flask.
When the gas has passed for a few minutes, cork the flask and allow it to
stand with a piece of white paper behind. Observe the slow appearance
of the yellow colour, due to NO2, indicating the time required for the
oxidation of NO in dilute gases (cf. p. 574).

Nitrogen dioxide, N02, and nitrogen tetroxide, N204. — If nitric
oxide is mixed with oxygen, or a gas containing free oxygen, red
fumes are produced. These consist of nitrogen dioxide : 2NO -f- 02
= 2N02- At temperatures below 140° a portion of the nitrogen
dioxide is associated, to form nitrogen tetroxide : 2N02 ^± N2O4.

If a mixture of 1 vol. of oxygen and 2 vols. of nitric oxide, both
gases being dry, is passed slowly through a long tube, so as to allow
time for complete oxidation, and the gas then passed through a
spiral tube cooled in a freezing mixture, a yellow liquid is condensed,
which is nitrogen tetroxide. But the reaction 2NO + 02 = 2N02
requires an appreciable time for its completion, and if the mixed «gas
is passed rapidly into a cooled tube, a green liquid condenses. This is
a mixture of nitrogen tetroxide and blue nitrogen trioxide formed
from the dioxide and unchanged nitric oxide. If the gases are
moist the liquid is always green : 4N02 + H2O = 2HN03 + N203.

Nitrogen dioxide is produced by the action of concentrated nitric
acid on copper or bismuth (Priestley) : Cu -f- 4HN03 = Cu(N03)2
-f- 2N02 + 2H20. It is obtained by the decomposition of lead and
copper nitrates by heat : 2Pb(N03)2 = 2PbO + 4N02 + 02.

EXPT. 239. — Heat dry
powdered lead nitrate in a
hard glass tube or retort, and
pass the red gas through a U-
tube cooled in a mixture of ice
and salt (Fig. 305). A yellow
liquid collects in the cooled
tube. Hold a glowing chip
over the exit of the U -tube :
it bursts into flame, showing
that oxygen is also evolved.
Pour the N2O4 on crushed
ice in a test-tube. A deep
blue layer of N2O3 separates : N2O4 + H2O ^±HNO2 + HNO3 ; 2HNO2
^± N2O3 -f H2O. On warming, red fumes are evolved (Fritzsche).

This is a very unsatisfactory method of preparing nitrogen dioxide
in quantity. It is more conveniently prepared by the action of
nitric acid and phosphorus pentoxide on a mixture of nitrous anhy-
dride and nitrogen dioxide obtained by distilling arsenious oxide with

-Ice

FIG. 305.— Preparation of Nitrogen Dioxide by
Heating Lead Nitrate.

xxix THE OXIDES AND OXY-ACIDS OF NITROGEN 591

a mixture of concentrated nitric acid and half its weight of concen-
trated sulphuric acid (Cundall, 1891) : N2O3 + 2HN03 ^=r 2N204 +
H2O.

EXPT. 240. — To the blue liquid obtained by distilling As2O3 with
nitric acid, and condensing in a freezing mixture, add excess of
P2O5, and fuming nitric acid drop by drop until the colour changes to
yellow. The mixture should be kept well cooled during the reaction.
Distil off through a worm cooled in ice, rejecting the first few c.c., which
are coloured green. Collect in a tube immersed in ice, and seal off.

The most convenient method is to heat nitrososulphuric acid with
sodium nitrate :

S02(OH)-ONO -f NaN03 = N204 + NaHS04.

EXPT. 241. — Pass sulphur dioxide into cooled fuming nitric acid
until . the liquid becomes a pasty mass of crystals of nitrososulphuric
acid. Add dry sodium nitrate. Warm, and collect the N2O4 as above.

Properties of nitrogen dioxide. — Nitrogen di- (or tetr-) oxide in a
good freezing mixture solidifies to pale yellow, nearly colourless
crystals, melting at —10-95° to a honey-yellow liquid. The solid
probably consists almost entirely of N2O4, which appears to be
colourless. The liquid at the melting point already contains a
trace of NO2, which is strongly coloured. On warming, the colour
of the liquid deepens ; at 10° it is distinctly yellow, at 15° it is
orange, and the colour darkens until at 21 -6° it is reddish-brown, and
then the liquid boils, giving a reddish-brown vapour. The colour of
the vapour also darkens on further heating, as may be seen by com-
paring two globes containing it, one maintained at the ordinary
temperature : at 40° the vapour has a very deep, almost black,
colour.

The colour change on heating is accompanied by a decrease in
vapour density up to 140°, when the density becomes constant, and
corresponds with N02 : the intermediate densities correspond with
the dissociation : N204 ^± 2N02, and the percentage dissociation, or
the percentage of N02 molecules in the vapour, may be calculated
by the formula given on p. 153.

Vapour density. Percentage NO2

Temperature A (H = 1) in vapour, by volume

26-7D 38-3 20-00

60-2 30-1 52-04

100-1 24-3 89-23

135-0 23-1 98-69

140-0 22-96 100-00

If the vapour is heated above 140°, the density further decreases,
but the colour becomes paler, until at 620° the gas is again colour-

592 INORGANIC CHEMISTRY CHAP.

less. This corresponds with the dissociation : 2NO2 ^ 2NO -f- O2,
which is complete at 620°. Recombination occurs on cooling, the
series of changes being passed through in the reverse order :

N2O4 solid ^± N2O4 liq. ^± N204 (vap.) =± 2NO2 ^± 2NO + Oa.

- 11° 26° 140C 620°

At 60-2° the vapour density is 30-1. The (theoretical) vapour
density of N2O4 is 46, that of NO2 is 23

.'. degree of dissociation y = — ^o— = ob ~ = ^'691.

This is the fraction by volume of the vapour consisting of NO2. The
fraction of N2O4 is 1-000 — 0-691 = 0-309, and since this corresponds
with 2 x 0-309 = 0-618 vol. of NO2, the fraction of NO2 by weight is

0-691 _

0-691 +0-618

The action of water on nitrogen dioxide has already been described
(p. 574). The composition of the gas is ascertained by passing
it over red-hot copper : 4Cu + 2N02 — 4CuO -f N2.

Nitrogen dioxide vapour does not support the combustion of a
taper, but strongly burning phosphorus and carbon burn in it.
The gas is probably decomposed by the temperature of the flame
into nitrogen and oxygen, or nitric oxide and oxygen. Potassium
inflames spontaneously in the gas ; heated sodium burns in it ;
and a spiral of iron wire heated to 500° also combines with the oxygen,
leaving half . the volume of nitrogen : 2N02 = N2 + 202. The
composition of the gas may be determined in this way. Tin is
oxidised to the dioxide, carbon monoxide to the dioxide at the
ordinary temperature ; hydrogen sulphide deposits sulphur, and
the nitrogen dioxide is reduced to nitric oxide : N02 + H2S =
NO + H20 + S. A mixture of the gas and hydrogen is reduced
to ammonia on passing over platinum sponge.

Nitrogen dioxide is absorbed by concentrated sulphuric acid with
formation of nitrososulphuric acid and nitric acid : since these
substances decompose each other, a state of equilibrium is attained :
N2O44-H2S04^±S02(OH)-0-NO + HN03. The gas is absorbed
by alkalies with formation of a mixture of nitrite and nitrate :
2KOH + N204 = KN02 + KN03 + H20. Baryta becomes in-
candescent at 200° in the gas : 2BaO + 2N204 = Ba(N02)2 +
Ba(N03)2. Quicklime, and oxides of zinc, aluminium, and lead,
absorb the gas on heating, but free nitrogen, to the extent of 30
per cent, of the N02, is liberated : 4CaO + 5N204 = 4Ca(N03)2
-f N2. Nitrites are also formed.

By passing nitrogen dioxide over finely -divided reduced copper,
nickel, or iron in the cold, Sabatier and Senderens (1893) obtained com-
pounds called nitroxyls. Copper nitroxyl is a brown substance, of the

xxix THE OXIDES AND OXY-ACIDS OF NITROGEN 593

formula Cu(NO2)2, decomposed at 90° : Cu(NO2)2 ^± Cu + 2NO2, and
by water, with formation of nitrate, copper, and nitric oxide.

The formula of nitrogen dioxide is probably O=N=O ; that

of the tetroxide ^N— N/ , although Divers regards it as a

true peroxide : 0:N-OON:0, and 0:N'0.

Pernitric acid- — Hautefeuille and Chappuis, and Berthelot (1881),
claimed to have obtained a higher oxide, N2O6 or N2O7, by the action
of a silent discharge or a mixture of nitrogen and oxygen : with water
it was supposed to form pernitric acid, HNO4. The existence of these
substances is highly doubtful.

Hyponitrous acid, H2N202. — Divers (1871), by reducing a solution
of sodium nitrite or nitrate with sodium amalgam, obtained a liquid
which, after neutralisation with acetic acid, gave a yellow precipitate
with silver nitrate. This had the empirical formula AgNO, and was
regarded by its discoverer as the salt of hyponitrous acid. Subse-
quent investigations showed that the acid really had the doubled
formula H2N202.

Sodium hyponitrite, Na2N202, is easily prepared by Divers'
process. Excess of sodium amalgam is added to a solution of sodium
nitrite : the reaction evolves heat, and by the prolonged action of the
amalgam any hydroxylamine formed is removed. The resulting
ammonia is removed by exposing the solution to concentrated
sulphuric acid in a vacuum desiccator. Granular crystals of sodium
hyponitrite, Na2N202,5H20, slowly separate. They are washed with
alcohol, and again exposed in the vacuum desiccator, when they fall
to a white powder of anhydrous salt, Na2N202, stable in air.

The nitrite is supposed to be reduced to the sodium compound
of dihydroxylamine, NH(OH)2, which is decomposed by the alkali :

2Na + 2H20 + NaN02 = NaN(OH)2 + 2NaOH
2NaN(OH)2 = Na2N202 + 2H20.

Hyponitrous acid is also formed in small quantities by the action
of nitrous acid on hydroxylamine :

HO - NjH2 "+ pi : N-OH = HONrN-OH + H2O.

EXPT. 242. — To a solution of hydroxylamine hydrochloride add
sodium nitrite and acetic acid. Heat rapidly to 60°, then add silver
nitrate solution. A yellow precipitate of silver hyponitrite is formed.

Sodium hyponitrite is most conveniently prepared by boiling
sodium hydroxylamine sulphonate (p. 553) with caustic soda :
2HONH-S03Na + 4NaOH = Na2N2O2 + 2Na2S03 + 4H20.

If silver hyponitrite is added gradually to an ethereal solution of

Q Q

594 INORGANIC CHEMISTRY CHAP.

hydrogen chloride in absence of moisture, and the filtered solution
evaporated in vacuo, crystalline explosive laminae of free hyponitrous
acid, H2N2O.2, are formed. The solution decomposes on heating with
evolution of nitrous oxide : H2N2O2 = H20 + N20.

EXPT. 243. — Warm a little sodium hyponi trite with dilute sulphuric
acid. Nitrous oxide is evolved with effervescence, and kindles a
glowing chip.

Hyponitrites in acid solution reduce permanganate : 5H2N2O2 -f-
8KMnO4 + 12H2S04 = lOHNOg + 4K2S04 + 8MnS04 + 12H2O.
In alkaline solution a nitrite is formed.

The doubled formula of the acid is supported by the following
evidence :

1. Acid and normal salts are known : KHN2O2 and K2N2O2. The
neutral point on titration is reached with KHN2O2.

2. The freezing point of the solution of the acid corresponds with
H2N202.

3. By the action of ethyl iodide on silver hyponitrite, ethyl hypo-
nitrite is obtained, the vapour density of which corresponds with the
formula (C2H5)2N2O2.

4. By oxidising hydroxylamine with silver oxide, hyponitrous acid,
an intermediate acid, nitrohydroxylamic acid, H2N2O3. and finally
nitrous acid, are obtained :

2NH2-OH -> H2N2O2 -» H2N2O3 -> H2N2O4(2HNO2).
Nitrohydroxylamic acid. — If methyl nitrate is added to a solution of
free hydroxylamine and caustic soda in methyl alcohol, the sodium salt
of nitrohydroxylamic acid, Na2N2O3, is obtained. This is very readily
oxidised by the air, with formation of nitfrite and nitrate, and is decom-
posed by boiling with water : 2Na2N2O3 + H2O = 2NaNO2 + N2O +
2NaOH. When the solid salt is gently heated, it decomposes into
nitrite and hyponitrite. On acidifying, the free acid liberated at once
decomposes into nitric oxide and water : H2N2O3 = 2NO + H2O.
The constitution of the acid appears to be HON:NO2H.

Nitrosyl chloride, NOC1.— The chloride of nitrous acid, nitrosyl
chloride, NOC1, is formed when nitric and hydrochloric acids are
mixed : HNO3 + 3HC1 = NOC1 + C12 -f 2H2O. A mixture of
1 vol. of concentrated nitric acid and 3 vols. of concentrated hydro-
chloric acid was called by the alchemists aqua regia because it is
capable of dissolving gold (" the king of metals "). It owes this
action to the presence of free chlorine. On warming aqua regia,
an orange-yellow gas is evolved, which is a mixture of chlorine and
nitrosyl chloride (Gay-Lussac, 1848). If the gas is dried by calcium
chloride and passed through concentrated sulphuric acid, the chlorine
passes on, whilst the nitrosyl chloride is absorbed as nitrososul-

xxix THE OXIDES AND OXY-ACIDS OF NITROGEN 595

phuric acid : NOC1 + SO2(OH)2 = S02(OH)-ONO -f HC1. If the
liquid is dropped on sodium chloride, and warmed, pure nitrosvl
chloride is evolved : S02(OH)-ONO + NaCl = S02(OH)ONa +
NOCL

Nitrosyl chloride is also formed by the direct combination of nitric
oxide and chlorine in bright sunlight, or in presence of animal char-
coal at 40-50° : 2NO + C12 = 2NOC1. Since it is the acid
chloride of nitrous acid, it is also formed by the action of phosphorus
pentachloride on potassium nitrite : PC15 + KNO2 = NOC1 +
POC13 + KC1.

Nitrosyl chloride is an orange-yellow gas with a suffocating odour,
easily condensed in a freezing mixture to a ruby-red liquid, b.-pt.
5-6°, freezing in liquid air to a lemon -yellow solid, m.-pt. — 60°.
It is readily decomposed by water and alkalies, in the normal manner :
NOC1 + 2KOH = KNO2 + KC1 + H2O. It has no action on gold
or platinum, but attacks mercury : Hg + NOC1 = HgCl + NO,
and most other metals. It is stable up to 700°, but then dissociates :
2NOC1 :^± 2NO -f- C12. It forms compounds with many metallic
chlorides, e.g., ZnCl2,NOCl ; FeCl3,NOCl, and is used in organic
chemistry, since it readily adds on to ethylene linkages :

\ c=C / + NOC1 -> \C(NO)-C1C/

Nitrosyl bromide, NOBr, a blackish-brown liquid, b.-pt. — 2°, is
formed by passing nitric oxide into bromine at — 15d. At the ordinary
temperature NOBr,Br2 is formed. Nitrosyl fluoride, NOF, is a gas,
b.-pt. — 56°, m.-pt. — 134°, formed by the reaction NOC1 + AgF =
NOF + AgCl. Nitrosyl perchlorate, NOC1O4, is formed by passing
NO -f- NO2 into very concentrated perchloric acid.

The chloride of nitric acid, NO2C1, is unknown, but nitryl fluoride,
NO2F, is formed by the reaction 4NO + F2 = 2NO2F + N2, at the
temperature of liquid oxygen. It is a gas, b.-pt. — 63-5°, m.-pt.
- 139°.

Nitrosyl sulphate, nitrososulphurie acid, or " chamber crystals,"
NO'HS04. — This important compound, which is supposed to be
formed as an intermediate stage in the lead chamber process
(p. 505), can be obtained in a number of ways.

It was obtained by Clement and Desormes by the interaction of
red fumes of oxides of nitrogen, sulphur dioxide, and a regulated
amount of moisture :

NO + N02 + S02 -f H20 = 2S02(OH)- 0-NO.
It is more conveniently prepared by passing the red vapours
from arsenic trioxide and nitric acid (p. 587) into concentrated
sulphuric acid : the acid soon deposits crystals of the compound :
:NO T N07=E* i N203 + H2S04 ^ S02(OH)-0'NO + H20. These

Q Q2

596 INORGANIC CHEMISTRY CHAP.

decompose with effervescence, evolving red fumes, when treated with
water, so that the reaction is reversible. They dissolve in concen-
trated sulphuric acid, and in sulphuric acid containing not more
than 35 per cent, of water, but if the acid is diluted below 65 per
cent. H2S04, decomposition occurs, and the nitrogen compounds
are then almost completely expelled.

EXPT. 244. — Pass red fumes from arsenious oxide and nitric acid
(p. 587) into cooled concentrated sulphuric acid. Observe the formation
of colourless crystals of nitrososulphuric acid. Pour off the liquid acid
and dilute with 40 per cent, sulphuric acid. Observe the colour change :
yellow, green, blue, and the sudden effervescence at a certain dilution.

Nitrososulphuric acid may also be prepared by passing sulphur
dioxide into cooled fuming nitric acid :

2S02 + 2HN03 = 2S02(OH)-0-NO.
The crystals are drained on a porous tile.

Nitrosyl sulphate, NO'HS04, is nitrososulphuric acid,
NO'0'S02(OH).

It is sometimes called nitrosulphonic acid, and its formula written
N02 S02-OH. The crystals melt with decomposition at 73° ; water
is split off, and dinitropyrosulphuric acid, S205(0*NO)2, formed :

so /°'NO JQ NO

2 0-NO

\0-NO

This is a white crystalline substance, m.-pt. 217°, b.-pt. 360°, also
obtained by passing nitrogen dioxide into liquid sulphur dioxide :
2S02 + 3N02 = (NO)2S2O7 + NO, or by heating the white solid,
oxynitrososulphuric anhydride, (N02'S03)2, obtained by the direct com-
bination of nitrogen dioxide and sulphur trioxide : (N02'S03)2 =
(NO)2S207 + O.

Nitrososulphuryl chloride, Cl-SOa'O-NO, is a crystalline solid, formed
by the direct combination of sulphur trioxide and nitrosyl chloride, or
by the action of thionyl chloride on silver nitrate.

Nitrogen sulphides. — Nitrogen sulphide, N4S4, is an orange-red crystal-
line solid, obtained by the action of dry ammonia on a solution of sulphur
chloride and chlorine, or on thionyl chloride. It decomposes at 185°,
is explosive on percussion, and is decomposed by cold water. It
combines with chlorine to form a tetrachloride, N4S4C14, and reacts
with S2C12 to form thiazyl chloride, N3S4C1, which is converted by
nitric acid into a crystalline nitrate, N3S4-NO3. The molecular weight

xxix THE OXIDES AND OXY-ACIDS OF NITROGEN 597

of nitrogen sulphide in solution corresponds with the formula N4S4 ;

^N-S;N

it is supposed to have the constitution SiSf' . Nitrogen

pentasulphide, N2S5, is formed, as a deep red liquid, when N4S4 is
treated with carbon disulphide at 100°. It decomposes on heating.

Sulphonic acids of ammonia and hydroxylamine. — Some products of
the action of nitrites on sulphites have already been described (p. 553).

The first product appears to be hydroxylamine disulphonic acid :
HO-NO + 2H-S03H = HO-N(SO3H)2 + H2O.

This may undergo hydrolysis or further sulphonation :

HON(S03H)2 + H20 = HO-NH-S03H (hydroxylamine sulphonic
acid) + H2S04.

HO-N(SO3H)2 + H-SO3H = N(SO3H)3 (nitrilosulphonic acid) + H2O.
These substances are intermediate products in the oxidation of sul-
phurous to sulphuric acid by means of nitrous acid.

N(SO3H)3 is a derivative of ammonia ; by boiling its salts for a short
time with water, they form salts of imidodisulphonic acid, NH(SO3H)2 :
N(S03K)3 + H20 = NH(S03H)2 + KHSO4. On further hydrolysis,
salts of amidosulphonic acid, NH2'SO3H, are formed. Sulphamide,
SO2(NH2)2, and sulphimide, (SO2NH)3, derived from sulphuric acid,
SO2(OH)2, are formed by the action of ammonia on a solution of
sulphuryl chloride, SO2C12, in benzene. Many other compounds of the
above types are known.

EXERCISES ON CHAPTER XXIX

1. What oxides of nitrogen may be obtained by the action of nitric
acid on metals ? How are the reactions explained ?

2. How is nitric acid manufactured from the atmosphere ? Describe
the reactions which take place in each stage of the process.

3. How is nitric acid obtained from (a) saltpetre, (b) ammonia ?
How may it be reconverted into ammonia ?

4. Describe the preparation of the following compounds in a state
of purity : (a) nitric oxide, (b) nitrous oxide, (c) nitrogen dioxide,
(d) nitrosyl chloride. What are their properties ?

5. How is nitrous anhydride prepared ? What is the action of the
substance on (a) water, (b) caustic soda, (c) concentrated sulphuric acid ?

6. How may the composition of (a) nitrous oxide, (b) nitric oxide,
(c) nitrogen dioxide, be found ?

7. Give examples of the oxidising and reducing reactions of nitrous
acid. What evidence is there as to the structural formula of this sub-
stance ?

8. How is hyponitrous acid obtained ? What is its structural
formula ?

9. How is nitrososulphuric acid prepared ? What is the action of
(a) water, (b) nitric acid, on this substance ?

CHAPTER XXX

THE INACTIVE ELEMENTS

The ratio of specific heats of a gas. — If 1 gm. mol. of a gas is
heated at constant volume from T° to (T -f- 1)° abs., the heat
absorbed is called the molecular heat at constant volume, Cv = Mcv,
where M = molecular weight, cv = specific heat at constant volume.
When a gas is heated at a constant pressure of 1 atm. it expands,
doing work against the atmospheric pressure, and the heat absorbed
is called the molecular heat at constant pressure, Cp = Mcp.

If the gas is ideal, i.e., obeys the law : pv — RT, no heat absorp-
tion results from the change of volume alone (cf. p. 258), and the
difference of molecular heats, (CP—CV), will be equal to the external
work done, viz. (pressure) X (increase of volume) :

= .#= 1-98 gm. cal.

In a monatomic gas the heat absorbed goes exclusively to increase
the kinetic energy of translation of the molecules (p. 262), and for

1° rise of temperature this increase of energy will be ^^ ( — ^- j .

gm. cal.

Hence, for a monatomic gas, Ctf = 297 gm. cal. But
Cp = C, -f- R — 4-95 gm. cal., hence the ratio of specific heats,

CP/CV, or Cp/cM usually denoted by y, is equal to -^- — 1-667 for

a monatomic gas.

If the gas molecule contains more than one atom, part of the
heat supplied at constant volume is used up in increasing the kinetic
energy of rotation of the molecule, considered as a rigid body ; in
addition, the energy of vibration of the atoms may be increased
if the molecule is not a rigid structure. If this extra energy is
denoted by E, per 1 ° rise of temperature, we shall have :

CH. XXX

THE INACTIVE ELEMENTS

599

which is. less than 1-667. It is found that CP/CV) for polyatomic
gases, is always less than 1-667, and is all the lower the more atoms
there are in the molecule :

GAS.

Helium...
Oxygen
Nitrogen
Air

Hydrogen
Carbon monoxide
Hydrogen
Chlorine

Even in the case of gases containing the same number of atoms in
the molecule (O2, C12 ; CO2, N2O ; SO2, H2S), the ratio y has different
values ; the lower values probably indicate the presence of additional
rotations or vibrations in the molecules to which they refer.

The value of CP/CV = y for a gas may be determined in two ways :
( 1 ) By allowing the gas, compressed to pressure p1 in a large globe, to
expand suddenly (adiabatic expansion) to atmospheric pressure, p2, by
opening a valve, and measuring the temperatures Tl and T2, before and
after the expansion, respectively, by means of a loop of fine platinum
wire (0-001 mm. diam.) placed in the centre of the globe, and used

as a resistance thermometer; pl/p2 = (Tl/T2)7^i. (2) From the

FOR- RATIO

FOB- RATIO

MULA. Cp/Cv.

GAS. MULA. Cja.

... He 1-667

Carbon dioxide ...

C02

•308 (0°)

... O2 1-398

Nitrous oxide

N20

•324 „

... N2

•402

Ammonia ...

NH3

•325 „

4N2 + 02

•403

Sulphur dioxide ...

SO2

•232 (20°)

•408

Hydrogen sulphide

H2S

•343 „

dde CO

•402

Methane

CH4

•313

)ride HC1

1-401

Ethylene

C2H4 1-255(0°)

... C12

•353

Steam

H20 1-33(100°)

velocity of sound in

the gas
RT

where p = pressure,

D = density. But p = , and D = M/v,

RT. „

M If P

p/D = RT/M,

is in dynes per cm.2, D in gm. per c.c.,
In corresponding units, R = 8-25 X 107.

u will be in cm. per sec.

A convenient method is to set up stationary waves in the gas, con-
tained in a thoroughly dry sealed tube clamped in the middle, and
brought into resonance by affixing discs of lead to the ends by sealing
wax (Fig. 306). The gas is caused to vibrate by stroking the tube with

FIG. 306.— Behn and Geiger's Method for Determining the Ratio of Specific Heats of a Gas.

a wet cloth, and the positions of the nodes and loops is indicated by
lycopodium powder, or silica dust, inside the tube. One end of the tube
is placed at the end of an open tube, also containing dust, the air in
which is caused to vibrate in resonance with the gas tube. The half
wave-length, X/2, is the distance from node to node, or loop to loop,
which is measured directly. If n is the frequency, u = nX, hence for

600 INORGANIC CHEMISTRY CHAP.

the gas and air, respectively, u^u^ = nX1/nA2 ; /. X12/X22 = ylMl/yzMz.
For air, yz — 1-403, and MI = 29, hence yl may be calculated.

The value y = 1-667 was found for mercury vapour by Kundt
and Warburg (1876), thus confirming the monatomic character of
the mercury molecule (p. 146).

This conclusion is necessary if the kinetic theory is accepted. The
assumption that the molecules of the inactive gases (argon, etc.), which
give the same ratio, are polyatomic, but that the atoms are " bound
by such enormous forces that they cannot be separated by chemical
affinity," therefore involves a denial of the kinetic theory — a contin-
gency which it must be assumed had not been realised. Any molecular
structure, no matter how rigid, which is not perfectly symmetrical
(i.e., monatomic) must possess rotational energy, and the maximum
value of CP/CV for such a molecule is 1-400.

THE INACTIVE GASES.

Argon. — In 1785, Cavendish, in his attempts to prove that the
nitrogen of the atmosphere is all of one kind, noticed that a small
residue was left on sparking with oxygen over caustic potash
(p. 566). Until 1894 it was taken for granted that atmospheric
nitrogen was homogeneous, but in that year Lord Rayleigh, in his
accurate determinations of the densities of gases (p. 71), noticed
that nitrogen prepared from the atmosphere is slightly heavier
than that prepared from oxides of nitrogen reduced by heated iron,
from ammonium nitrite, or from urea and sodium hypobromits :
Normal density : (a) " chemical " nitrogen = 1-25107 ; (6) atmo-
spheric nitrogen = 1-25718.

This difference did not escape such an accurate observer, and
a repetition of Cavendish's experiment confirmed the presence of a
small unabsorbed residue, which did not give the spectrum of
nitrogen.

In conjunction with Sir William Ramsay, Rayleigh now attempted
to prepare the new gas from atmospheric nitrogen in quantities
sufficient to permit of a careful examination of its properties.
Two methods were employed : (i) absorption of the nitrogen by
red-hot magnesium ; (ii) conversion of the nitrogen into nitric
acid by sparking with oxygen in presence of an alkali.

1. The oxygen of air was absorbed by passing over red-hot copper, and
the residual nitrogen then repeatedly passed over heated magnesium.
The nitrogen was slowly absorbed as magnesium nitride, Mg3N2, and
the unabsorbed residue was collected and examined. The apparatus
used is shown in Fig. 307. The atmospheric nitrogen, contained in a
gas-holder, A , was passed through drying tubes and then through a tube,
G, containing red-hot magnesium. The gas was collected in the gas-

xxx THE INACTIVE ELEMENTS 601

holder, B. It was then passed back again, and the process repeated
until no further absorption took place ; the volume of the gas was
reduced to l/80th. Further treatment raised the density of the gas to
19-94 (H = 1).

2. A mixture of 11 vols. of oxygen and 9 vols. of air was passed
(Fig. 308) into a 50-litre glass globe, provided with heavy platinum

SIR WILLIAM RAMSAY.

electrodes. A discharge from a transformer of 6000-8000 volts was
passed between the electrodes, and a fountain of caustic soda solution
discharged over the inside of the globe. With a consumption of energy
of 1 horse-power, 20 litres of gas were absorbed per hour. The oxygen
was absorbed from the residual gas by p^rogallol and alkali.

The new gas was distinguished from all other elements by its
entire inertness. It is not absorbed by heated metals, copper

602

INORGANIC CHEMISTRY

oxide, caustic potash, potassium permanganate, sodium peroxide,
phosphorus, etc., nor does it react when sparked with oxygen,

hydrogen, chlorine, or even fluor-
ine. It is unchanged when an arc
is maintained in the gaseous or
liquid substance for several hours.
On this account, Ramsay called
the gas argon (Greek argon, lazy,
or inactive). Berthelot, however,
in 1 895, stated that a contraction
occurred when a mixture of argon
and benzene vapour was sub-
jected to the silent discharge : this
is the sole experiment indicating
any activity of argon, and it is
unconfirmed.

The separation of atmospheric
argon is now carried out on the
technical scale, since the gas is in
demand for filling metal-filament
electric lamps. If these are
vacuous, the metal filament
(composed of tungsten) volatilises,
and a black film is deposited on
the inside of the bulb, which re-
duces the efficiency of the lamp
by obscuring the light. If the
lamp is filled with argon, the
blackening of the bulb is consider-
ably reduced. The argon is ob-
tained by circulating air through
a mixture of 90 parts of calcium
carbide and 10 parts of calcium
chloride, heated to 800° in iron
retorts. The nitrogen and oxygen
are absorbed, the latter as calcium
cyanamide (p. 544), the former as
calcium carbonate, and the resi-
dual gas, after passing over heated
copper oxide to oxidise carbon
monoxide (which is absorbed by
potash), is dried. Argon is also
obtained from the liquid oxygen
left after the evaporation of liquid
air, which contains about 3 per
cent, of argon. The oxygen is

XXX

THE INACTIVE ELEMENTS

603

To Transformer

removed from the gas obtained by evaporation, by passing over
heated copper, or by a hydrogen flame, and the residual gas freed
from nitrogen by heated carbide. Argon is also obtained by the f rac-
tionation of liquid air, e.g., in the Claude apparatus, and the gas, con-
taining about 87 per cent, of argon, used for filling lamps. Argon ob-
tained by all these processes
contains other inactive gases
(e.g., krypton) in traces.

Inactive gases are also
evolved from hot-springs
having their sources at great
depths in the earth. The
spring of Bourbon-Lancy
evolves 16,000 litres of in-
active gases per annum, of
which 10,000 litres are
helium (see below). The
water of these springs is
radioactive.

In the residue from the
evaporation of a large
volume of liquid air Ramsay
(1898) discovered two other
new inactive gases, krypton
(Greek krypton, concealed),
and xenon (Greek xenos, the
stranger). In crude liquid argon two other inactive gases, helium,
and neon (Greek neon, new), were found. The latter was searched
for in order to fill a gap in the periodic system (p. 471) between
helium (4) and argon (40). Since helium is the element of least
atomic weight in the group, the class of inactive elements may be
called the helium group. They are most easily characterised by their
spectra, in Geissler tubes. On prolonged exposure of the gas to
the discharge, the light emitted by the tube diminishes in intensity.
This is not due to absorption of the inert gas, by the electrodes, but
to removal of traces of nitrogen, etc., by the latter. In the absence
of traces of diatomic gases, the inactive gases become fluorescent,
or even non-conducting.

The examination of the residues from the evaporation of 120 tons
of liquid air failed to indicate the presence of any other gases than
those described.

Helium. — In 1868, the spectroscopic examination of the chromo-
sphere of the sun during a total eclipse revealed the existence of a
new yellow line, which did not exactly coincide with the D lines of
sodium. Janssen called this line D3, and Frankland and Lockyer
concluded that it corresponded with an element not present in

Gases in

FIG. 308.— Rayleigh's Method for the Preparation
of Argon.

604 INORGANIC CHEMISTRY CHAP.

terrestrial substances, to which they gave the name helium (Greek
helios, the sun). In 1894, Ramsay, at the suggestion of Miers,
examined the gas evolved from cleveite (a variety of pitchblende),
which had been supposed by Hillebrand (1888) to be nitrogen.
This gas is evolved by heating the mineral with dilute sulphuric
acid, or in a vacuum. It contains about 20 per cent, of nitrogen,
but when this is removed by sparking with oxygen over alkali,
there is a residue, which was found by Crookes to give, among other
lines, the D3 line in the spectrum. The gas was the unknown element
of Frankland and Lockyer.

Doubts having been cast on the homogeneity of the gas, Ramsay
and Travers (1897) showed, by an exhaustive fractional diffusion,
that it could be separated into a light fraction, showing all the pro-
perties of helium, and unaffected by further diffusion, and a heavier
fraction containing argon.

Helium was afterwards discovered in traces in the atmosphere,
in gases occluded in the rare mineral broggerite, in the gases of
mineral springs (Cauterets, Bath, etc.), and especially in the natural
gas from different localities in Kansas, U.S.A., some specimens of
which contain more than 1 per cent, by volume of helium. It
has been prepared in large quantities from this natural gas, and,
on account of its non-inflammable character, proposed for use in
filling airships.

Helium occurs in small quantities in numerous minerals, and
there is a good deal of evidence that its presence is the result of
radioactive changes which have taken place at remote periods (p. 1034).
The gas, although present only in minute quantities in the atmo-
sphere (1 vol. in 250,000), may be separated by a slight modification
of the Claude rectifier (p. 177) for the treatment of liquid air.

Helium is readily purified from other gases by making use of
the discovery of Dewar (1904) that cocoanut charcoal at the
temperature of liquid air completely absorbs all gases except
hydrogen, helium, and neon. Quartz at a temperature of 1100°
is permeable only to hydrogen and helium.

Liquid helium was first obtained by Kamerlingh Onnes in 1907,
by the free expansion of the gas, previously cooled to 15° abs. ;
"me liquid has a density of only 0-122, and has a very flat men-
iscus, indicating a small surface-tension. It boils at 4-3° abs. ;
by the rapid evaporation of the liquid, solid helium was obtained,
and the temperature reduced to 1-5° above the absolute zero. At
this temperature the electrical resistance of metals practically
vanishes, so that a current set up by magnetic induction in a
closed ring of the metal, cooled in liquid helium, continues to cir-
culate for several days.

Other inactive gases. — Neon occurs in traces in the air(l vol. in 55,000);
it is separated by fractionation, but more readily by Dewar's method.

THE INACTIVE ELEMENTS 605

The inert gases are brought in contact with charcoal at — 100°, when
the argon, krypton, and xenon are completely absorbed. The residual
helium and neon are pumped off, and brought in contact with charcoal
cooled to — 185° ; the neon is absorbed, and the helium (with a little
neon) tan be pumped off. On warming the charcoal, the neon is
expelled. If the first charcoal bulb is now warmed to — 80°, pure
krypton is evolved ; at higher temperatures, a mixture of krypton and
xenon comes off. This gas is recondensed on charcoal at —150°,
and the bulb put in connection with a second charcoal bulb cooled to
— 180° ; the krypton passes over, leaving xenon in the first bulb. The
gases are separated from the respective bulbs by warming (Valentiner
and Schmidt, 1905). In a Geissler tube (p. 193), neon gives a beautiful
orange -pink light ; the same light is seen if a tube of neon at atmospheric
or lower pressure, containing mercury, is shaken in a dark room (Collie).
Neon is obtained from the residues of the Claude air liquefiers ; a
machine making 50 cu. m. of oxygen per hour produces 100 litres of
neon per day. The gas is used for filling electrodeless vacuum lamps.

Another element of the group of inactive gases is niton, the emanation
of radium : this will be considered in Chapter LI.

Properties of the helium group. — Since the inactive elements are
devoid of all chemical affinities, they are completely described
by an enumeration of their physical properties, given in the following
table. Although niton may appear to be an intensely active
substance, this is really due to its atomic disintegration ; in itself
it is a perfectly inert gas:

Helium.

Neon.

Argon.

Krypton.

Xenon.

Niton.

Normal den-
sity
Atomicweight
(H = 1) . .
Critical tem-
perature
(abs.). .

0-1786
3-97

5°

0-9002
20-0

60°

1-7818
39-6

150-6°

3-708
82-26

210-5°

5-851
129-2

289-6°

9-97
220-6

377-5°

Critical pres-
sure (atm.)
Boiling point
(abs )

2-75
4-5°

29
27-1°

47-97

86°

54-3
122°

58-2
163-9°

62-5
211°

Melting point
(abs )

83-4°

104°

133°

202°

Compressibi-
lity (p. 148)
A b s o r ption
c o e ff . in
water at 0°
Ratio of speci-
fic heats, 7.

±0

0-0134
1-652

-0-00105

0-0114
1-642

+ 0-00081

0-0561
1-65

+0-00210

0-1207
1-689

+ 0-00690

0-2189
1-666

0-5

606 INORGANIC CHEMISTRY CH. xxx

A uniform gradation is apparent in many of these properties ;
this, however, is broken in the case of neon with the compressibility,
and solubility. The monatomic character of the gases is indicated
by the values of the ratio of the specific heats, and is confirmed
by other lines of evidence (e.g., the refractive indices).

The inactive gases form a separate group in the Periodic System,
and in conformity with the rule of valency (p. 463), this is called
the zero-group, Group 0. They bridge the gap between the
strongly electropositive elements of the first group and the strongly
electronegative elements of the seventh group (p. 471).

EXERCISES ON CHAPTER XXX

1. Give a brief account of the history of the discovery of the inactive
gases. How is argon obtained from the air, and for what purpose is
it used ?

2. From what sources may helium be obtained ? What possible
use has been suggested for this gas ?

3. How has the monatomic character of the inactive gases been
established ? What is their position in the periodic table ?

4. Describe briefly how the different inactive gases may be separated
from one another.

CHAPTER XXXI

PHOSPHORUS

The nitrogen group. — Group V. in the Periodic System comprises,
besides radio-elements (Chapter LI), the following elements :

Odd series : nitrogen, phosphorus, arsenic, antimony, bismuth.

Even series : vanadium, niobium, tantalum.

Of these, all except nitrogen, phosphorus, and arsenic are metals:
the non-metal nitrogen has been dealt with, and the other two
non-metals, phosphorus and arsenic, are discussed in the present
and following chapters. The metals are considered later.

The properties of the elements of the odd series are as follows :

Atomic weight (H =
Sp. gr. of solid

N.
1) 13-897

0-79

P. As.

.30-79 74-37
I 1-83 )

Sb. Bi.

119-2 206-4

•
Atomic volume . .

(liq.)
17-61

< (yellow) V 5-73
/ 2-20(red) \
16-69 12-98

6-62 9-80
18-02 21-08

Melting point . . . .

— 210°

(yellow)
44° 850°

630-0° 271°

Boiling point .

-195-7°

287° 450°

1440° 1420°

sublimes

The typical compounds of the elements, in which the latter are
usually ter- or quinque-valent, but occasionally quadrivalent, are
as follows :

NH3, N2H4, N3H
NC13
N20, NO, N406, N02,
N204, N205

PH3, P2H4, P12H6, P9H2
PC13, PC15
P406, P204, P4010

AsH3

AsCl3, AsCl5 (?)
As4O6, As2O5

SbH3
SbCl3, SbCl5
Sb4O6, Sb2O4, Sb2O5

BiH3(?)
BiCl3
Bi203, Bi204, Bi205

The hydrides of these elements are all gaseous. Ammonia is a
relatively strong base ; phosphine (PH3) is a very weak base,

607

608 INORGANIC CHEMISTRY CHAP.

whilst arsine (AsH3) and stibine (SbH3) are devoid of basic pro-
perties. Bismuth forms a very unstable gaseous hydride, which
dissolves in solutions of alkalies, and may be feebly acidic. The
oxides of nitrogen are more numerous than those of the other
elements, of which the types R2O3, R2O5, and sometimes R2O4
only are known. The acidic character of these oxides, i.e., the
electronegative character of the elements, diminishes from nitrogen
to bismuth ; from arsenic onwards the oxides also show basic
properties : stable salts derived from Sb2O3, and Bi203, e.g.,
Bi(N03)3, are known. The halogen compounds of phosphorus are
completely hydrolysed by water : PC13 + 3H2O = H3P03 + 3HC1 ;
those of arsenic can exist in presence of excess of acid : AsCl3 -f 3H20
^ H3As03 + 3HC1 ; those of antimony and bismuth are only
partially hydrolysed : BiCl3 + H2O = BiOCl + 2HC1.

Phosphorus. — About 1669, a physician of Hamburg, Brand,
obtained a remarkable substance by distilling evaporated urine
with sand and charcoal. It had the property of shining, the glow
being visible in the dark, and was called phosphorus (Greek phos,
light, and phero, I bear). Urine contains microcosmic salt,
NaNH4HP04 ; on heating, this yields sodium metaphosphate, NaP03,
which is reduced on ignition with charcoal : 2NaP03 + 40 =
Na2C03 + 2P + 300. The secret of the process was sold by Brand
to Krafft ; the latter exhibited the product in the Court of Charles
II in 1677. Here it was seen by Boyle. The latter, and Kunckel
in Berlin, independently rediscovered the method of preparation
in the year 1@78. Boyle called the substance the noctiluca, but it
was generally known as " Boyle's," or " English," " phosphorus "
to distinguish it from the Bolognian phosphorus (BaS, p. 877), which
emitted a similar light, but only after previous exposure to sunlight.
Scheele, in 1770, discovered calcium phosphate, Ca3(P04)2, in
bones, and Gahn prepared phosphorus from bone-ash. The pro-
cess formerly in use on the large scale (see below) for the preparation
of phosphorus from bone -ash was devised by Scheele. The
elementary nature of phosphorus was recognised by Lavoisier in
1777.

Occurrence of phosphorus. — Phosphorus occurs always in the
combined state. The primary mineral appears to be apatite,
3Ca3(P04)2,CaF2 ; chlorapatite, 3Ca3(P04)2,Ca012, also occurs. These
are hard minerals, practically insoluble in dilute acids. From
them, by weathering, the secondary deposits of phosphates have
probably been formed, although many of these consist of fossil
bones, in the formation of which the phosphates were first assimilated
by animals. The so-called " soft phosphates " are coprolites (calcium
phosphate of fossil excreta) and Charleston phosphate (27 per cent.
P205), from river beds in South Carolina, and are easily decomposed
by sulphuric acid. " Hard " varieties are estramadurite (33 per cent.

xxxr PHOSPHORUS 000

P205), sombrerite (35 per cent. P206), which are Spanish minerals,
and Redonda phosphate (35-40 per cent. P205), a cheap and rich
ore from the West Indies. The softer ores are used in the manu-
facture of phosphorus by the old process (q.v.), or of superphosphate ;
the hard varieties can be used in the modern electric furnace process
for the preparation of the element. Phosphorus is an essential
constituent of vegetable and animal tissues, occurring especially
in the seeds, in the yolk of eggs, in the nerves and brain, and in
bone-marrow, usually in the form of fats containing esters of phos-
phoric acid, known as lecithins, or glycerophosphates.

In the processes of tissue-metabolism, the organic phosphoric
esters (lecithins] are broken up, and the phosphoric acid is excreted,
through the agency of the kidneys, in the form of microcosmic salt.
In order to repair the tissue-waste and to provide phosphates for
the structure of bones, phosphorus compounds must form essential
constituents of foods. Plants take up the element from the soil
in the form of calcium phosphate, which dissolves in water con-
taining carbonic acid. Phosphates, such as bone-meal, or super-
phosphates (p. 849), are therefore valuable fertilisers. The natural
phosphates in the soil are probably derived from the weathering of
apatite.

Phosphorus occurs in an inorganic form in the bones, which in the
fresh condition contain about 58 per cent, of calcium phosphate,
Ca3(P04)2, together with some calcium carbonate, fats, and organic
matter containing nitrogen. The fat is extracted by solvents such
as carbon disulphide, or chlorinated acetylenes (p. 680), and when
the degreased bones are boiled with water under pressure
in autoclaves, much of the remaining organic matter is dis-
solved. On evaporating the solution, glue is obtained. If the
solid residue of the bones is now heated strongly out of contact
with air, in iron retorts, the remaining organic matter is decom-
posed, and animal charcoal (p. 668) remains, which is used in
decolorising sugar syrup. When it is no longer active, it is calcined
in the air, when the carbonaceous matter is burnt off, and a white
mass of bone-ash is left, consisting of about 83 per cent, of
calcium phosphate, with calcium carbonate and a little fluoride.

Preparation of phosphorus. — Phosphorus was formerly prepared
by Scheele's process from bone-ash, or soft mineral phosphates.
These were decomposed by hot sulphuric acid (sp. gr. 1-5-1-6), so
as to form insoluble calcium sulphate and phosphoric acid. The
phosphoric acid solution was filtered off, evaporated to a syrup,
mixed with powdered coke, and distilled in fireclay retorts at a
white heat :

Ca3(P04)2 + 3HoS04 — 3CaS04 -f- 2H3P04 (orthophosphoric acid).
H3P04 = H20 -f HP03 (metaphosphoric acid ; formed on heating).
4HP03 + 12C = 2H2 -f- 12CO + P4 (phosphorus).

R R

610

INORGANIC CHEMISTRY

The phosphorus distilling over was condensed under water.

Practically all the phosphorus is now made by a method proposed
by Wohler (1829) in the electric furnace (Readman, Parker, and
Robinson process, 1888). This method is applicable to hard,
sparingly-soluble phosphates, since the mineral is not treated with
acid. A mixture of phosphate, sand, and coke is fed by a worm-
conveyor into a closed electric furnace, provided with an outlet
above for the gases and phosphorus vapour, a slag hole below, and
carbon electrodes between which an electric arc is struck (Fig. 309).
The phosphate is decomposed at the high temperature by silica,
which is very difficultly volatile and weakly acidic :

Ca3(P04)2 + 3Si02 = 3CaSiOa + P205 (at 1150°).
The calcium silicate forms the molten slag. The vapour of phos-
phorus pentoxide is reduced
by the carbon at about
1500°, forming carbon mon-
oxide and phosphorus
vapour, which pass out at
the top : PoO5 + 5C = 2P
+ SCO. About 5 kilowatt-
hours are used per gram of
phosphorus : the yield is
80-90 per cent. The cooled
gas is passed into water,
when crude phosphorus con-
denses as a dark-coloured
mass. It is purified by melt-
ing under a solution of
chromic acid, when some
of the impurities are oxidised
and pass into solution, and
others are separated and
rise as a scum. The liquid

phosphorus may also be filtered by pressing through chamois
leather. The colourless phosphorus is finally cast into wedges
(about 2 Ib.) in tin moulds, or into sticks, by running the liquid
into glass tubes cooled in water, and drawing out the stick at the
other end.

The annual production of phosphorus amounts to about 5,000 tons,
most of it being vised in the manufacture of matches. Some phosphorus
is used in making phosphor-bronze, as a poison for rats, and in the
preparation of phosphorus trichloride, pentachloride, and pentoxide,
in chemical industries and laboratories.

EXPT. 245. — Mix 1 gm. of powdered sodium metaphosphate (obtained
by heating microcosmic salt in a platinum crucible) with 0-5 gm. of'

FIG. 309.— Electric Furnace for Manufacture of
Phosphorus.

XXXI PHOSPHORUS (m

aluminium powder and 3 gm. of fine white sand. Heat the mixture
strongly in a hard glass tube in a current of dry hydrogen. Phosphorus
distils over, condensing in the cool part of the tube. White fumes with
a strong smell of phosphorus escape from the exit tube, which dips
under water.

White (or yellow) phosphorus.— Ordinary white phosphorus, made
as described, is a translucent white solid, like wax. It is
soft enough at the ordinary temperature to be cut with a kni'e —
an operation which should always be performed under water.
Phosphorus is kept in bottles under water on account of the ease
with which it takes fire in air. Below 5-5° the phosphorus becomes
brittle, and the crystalline structure produced on cooling may be
seen by etching the stick of phosphorus in concentrated nitric acid.

The specific gravity of white phosphorus is 1-82, and its melting
point under water is 43-3°. In dry glass tubes it melts at 30°.
The liquid exhibits supercooling. Phosphorus boils at 269° (various
temperatures, from 269° to 290°, have been recorded, the dis-
crepancies being probably due to the partial conversion of fused
phosphorus into the red variety above 200°), yielding a colourless
vapour, the density of which, between 512° and 1000°, corresponds
with the formula P4. Between 1500° and 1700° the density
decreases, indicating partial dissociation : P4 ±z 2P2. According
to Stock, the dissociation is 1 per cent, at 800°, and more than 50
per cent, at 1200°.

White phosphorus is very sparingly soluble in water (1 in 300,000),
but dissolves in benzene, turpentine, olive oil, sulphur chloride,
phosphorus trichloride, and especially in carbon disulphide (9 parts
of P in 1 part of CS2). From the elevation of the boiling point of the
latter solvent, Beckmann found the molecular formula P4, agreeing
with that of the vapour, and Hertz obtained the same result from
the depression of freezing point of benzene. On evaporation out
of contact with air, the solution in carbon disulphide deposits large,
transparent, regular crystals, usually rhombdodecahedra, which
exhibit a play of colours resembling that of the diamond. These
crystals may also be formed by the slow sublimation of phosphorus
in an evacuated tube, one end being kept cool by a moist cloth ;
the tube is preserved in the dark, since on exposure to light the
crystals become red and opaque. By shaking melted phosphorus
under a cold solution of urea, it is obtained in the form of a fine
powder. White phosphorus dissolves in cold concentrated nitric
acid, forming phosphoric acid (q.v.).

The characteristic property of white phosphorus is the ease
with which it undergoes oxidation when exposed to the air at the
ordinary tempera/ture, the spontaneous oxidation being accom-
panied with a green glow, or phosphorescence. If gently warmed

R R 2

012 INORGANIC CHEMISTRY CHAP.

to about 34°, it catches fire in dry air, and burns with a brilliant
white light, forming white fumes of the pentoxide, P205. Finely-
divided phosphorus takes fire spontaneously in the air. It may be
burnt under water in a current of oxygen.

EXPT. 246. — Place a few pieces of phosphorus in a test-tube supported
in a beaker of water. Half fill the test-tube with water, and pass through
a current of oxygen. Now heat the water in the beaker. When the
temperature reaches 60° the phosphorus takes fire and burns under
water where it comes in contact with the oxygen.

EXPT. 247. — Pour a solution of phosphorus in carbon disulphide on
a piece of blotting-paper supported on a tripod stand. The solvent
rapidly evaporates, and the finely -divided phosphorus left catches
fire and burns with the formation of fumes of P2O5. The paper is
charred, but does not burn, since phosphoric acid, formed from the
oxide by moisture in the air, is readily fusible, and protects the paper
from contact with the air. For the same reason it is difficult to ignite
a piece of paper in a phosphorus flame. The solution in ether exhibits
phosphorescence when poured on hot water, or rubbed on the skin.

Sticks of white phosphorus kept under water become covered
with a white crust, which may be an allotropic modification, or an
oxide, since, according to Baudrimont, it is not formed in .water
free from air. This crust slowly turns red, and finally black, and
the dark colour spreads through the mass of the phosphorus.

White phosphorus is very poisonous, the lethal dose being about
0-15 gm. Workmen exposed to the vapour are liable to decay of
the bones, especially of the jaw (" phossy-jaw "), and its use in the
manufacture of matches has been prohibited in many countries.

Red phosphorus. — This modification, formerly called "amor-
phous phosphorus," was prepared by Schrotter in 1845 by heat-
ing white phosphorus for a few hours at 250° in a flask filled with
nitrogen or carbon dioxide. The liquid deposits a red powder, and
finally solidifies to a purplish-red mass. The transformation begins
at about 230° ; it is fairly rapid at 250°, and at higher temperatures
becomes reversible. Considerable amounts of heat are evolved :
P (white) == P (red) +3-7 kgm. cal.

Red phosphorus is also left as a residue when white phosphorus
burns in air, or in oxygen under water, and was, until Schrotter's
discovery, considered to be a sub-oxide.

EXPT. 248. — Heat a little white phosphorus in a strong sealed glass
tube suspended by a wire in the vapour of diphenylamine boiling, at
310°, in a glass jacket (Fig. 310). The clear liquid deposits red phos-
phorus and slowly solidifies.

Brodie (1853) showed that the transformation of white into red

xxxi PHOSPHORUS (;13

phosphorus is considerably accelerated by the presence of a little
iodine, and then occurs at 200°. The same change occurs when
a little iodine, or selenium, is added to a solution of
white phosphorus in carbon disulphide.

Red phosphorus is manufactured by heating
about a ton of phosphorus in a large cast-iron
pot provided with a cover, through which passes
an upright iron tube about 6 ft. long and 1 in.
in diameter. The pot is carefully and uniformly
heated to 240-250°, the temperature of the fused
phosphorus being controlled by thermometers,
protected by iron tubes, since phosphorus attacks
glass. A little phosphorus burns, absorbing the
oxygen from the air in the vessel, and oxidation
then ceases- The hard solid left in the pot when

White into Red the conversion is complete is ground up under

water, and boiled with a solution of caustic soda

to free it from unchanged white phosphorus (p.

618). It is then repeatedly washed with hot water and dried

with steam. It contains about 0-5 per cent, of white phosphorus,

and some phosphoric acid.

Red phosphorus has a density of 2-25. It is not self-luminous,
has no taste or smell, and is not poisonous. On exposure to air,
very little change occurs, although slight oxidation takes place,
the dry powder becoming moist and phosphoric acid being formed.
The powder does not ignite in the air until heated to about 240°.
The melting point of red phosphorus is between 500° and 600° ;
when strongly heated it is converted into vapour, which on cooling
deposits white phosphorus.

EXPT. 249. — Place a small heap of red phosphorus near one end of
a flat piece of tinplate, and a small piece of white phosphorus at the
other end. Support the tinplate on a tripod stand, and heat the end
near the red phosphorus with a small Bunsen flame. The white phos-
phorus catches fire first, although it is further from the flame than
the red phosphorus. The latter has, therefore, a higher ignition point.

EXPT. 250. — Place a little red phosphorus in a hard glass test-tube,
fitted with a rubber stopper and two tubes. Displace the air from the
tube by a slow stream of carbon dioxide, and heat the phosphorus
strongly. Colourless drops of white phosphorus distil on to the cooler
portion of the tube.

White phosphorus appears to be an unstable form : it passes
slowly into red phosphorus, even at the ordinary temperature
when exposed to light. The vapour pressure of white phosphorus
at 200° is greater than that of red phosphorus at 350°, and if white
phosphorus is placed in one limb of a U-tube at 324°, and red

614 INORGANIC CHEMISTRY (HAP.

phosphorus at 350° in the other, distillation occurs from the cooler
to the hotter position.

Allotropic forms of phosphorus. — Besides ordinary white phos-
phorus, or a-white phosphorus, two other white forms have been
described. /3-white phosphorus is formed by cooling the a-form
to — 76-9°, or by subjecting it to about 12,000 atm. pressure : it
crystallises in the hexagonal system, Vernon described another
form, y- white phosphorus, m.-pt. 45-3°, sp. gr. 1-827, obtained in
rhombic prisms by cooling liquid phosphorus very slowly.

Scarlet phosphorus was obtained by Schenck by boiling a 10 per
cent, solution of white phosphorus in phosphorus tribromide for
ten hours. It deposits as a fine scarlet powder, more active than
red phosphorus, but differing from white phosphorus in not oxidising
in the air or being poisonous. For the latter reason, it is now used
in the manufacture of matches. It dissolves in alkalies, evolving
phosphine (q.v.), and turning dark in colour. Prepared as above, it
always contains tribromide, but may be obtained pure by heating the
tribromide with mercury at 240° : 2PBr3 + 3Hg = 3HgBr2 + 2P.

Metallic phosphorus, or a-black phosphorus, is formed (Hittorf,
1865) by heating ordinary red phosphorus in a sealed tube at 530°,
the upper portion of the tube being kept at 444°. Brilliant, opaque,
monoclinic. or rhombohedral, crystals, sp. gr. 2-316 or 2-34, which
do not oxidise in air, sublime. These crystals are also formed
by dissolving phosphorus in lead at 400° in a closed tube, allowing
it to crystallise, and dissolving out the lead with dilute nitric acid.
This modification is not a conductor of electricity.

/3-black phosphorus, sp. gr. 2-69, m.-pt. 587-5°, is formed irreversibly
from white phosphorus, at 200° under a pressure of 12,000 kgm. per
sq. cm. It does not ignite at 400° in air, and is a fairly good conductor
of electricity.

Violet phosphorus is formed by heating white phosphorus with a
trace of sodium to 200° under very high pressure. It is crystalline,
sp. gr. 2-35, m.-pt. 589-5°.

Red phosphorus was long considered to be amorphous, but Pedler
and Retgers, in 1890, showed that it consists of small rhombohedral
crystals. It is not considered to be a definite modification of phos-
phorus, since its properties (e.g., heat of combustion) are variable,
but is supposed to consist of a solid solution of scarlet phosphorus
in metallic phosphorus. Some white phosphorus may also be
present. It is insoluble in carbon disulphide, and is a feeble con-
ductor of electricity. Whereas white phosphorus ignites spon-
taneously in chlorine, red phosphorus burns in the gas only when
heated.

The glow of phosphorus. — The spontaneous oxidation of phos-
phorus, which takes place when white phosphorus is exposed to

xxxi PHOSPHORUS (iir,

air, is accompanied by the emission of a faint green glow, white
fumes being at the same time evolved. The glow is produced'
when only minute traces of phosphorus or oxygen are present,
and its formation is used as a test for free phosphorus, when the
latter is suspected in cases of poisoning.

EXPT. 251. — A small piece of phosphorus is added to water in a flask
connected with a Liebig's condenser (Fig. 311). On boiling the water,
the phosphorus distils over with the steam, and a phosphorescent glow
is seen in a dark room at the point in the condenser where the vapours
deposit liquid.

EXPT. 252. — The glow of phosphorus is strikingly shown in the
following experiment, due to Smithells (" the cold flame "). A few
pieces of phosphorus are placed in a receiver, which
is then filled up with glass wool. The receiver is
heated gradually on a water-bath, a stream of
carbon dioxide being passed through (Fig. 312).
The phosphorus vapour carried along with the

Jim

FIG. 311. — Detection of Phosphorus.

FIG. 312.— Smithells'
" Cold Flame."

gas oxidises in the air, and a green flame appears at the top of the exit
tube. This is so cool that the hand may be held in it, and it will not
kindle the head of a match.

The glow of phosphorus was investigated by Boyle, who found
that : (1) phosphorus glows only in the presence of air ; (2) an acid is
produced which differs from phosphoric acid, since it gives little
flashes of light on heating [phosphorous acid] ; (3) the glow is
exhibited by solutions of phosphorus in olive, and some other,
oils, but oils of mace and aniseed prevent it ; (4) a very small
quantity of phosphorus (1 part in 500,000 parts of water) can be
detected by the glow ; (5) after exposure to phosphorus, the air
acquires a strong odour [ozone], distinct from the visible fumes.

616

INORGANIC CHEMISTRY

CHAP.

Although a large number of other investigators have since
examined the glowing of phosphorus, it cannot be said that any great
advance has been made from the facts ascertained by Boyle in the
seventeenth century. A little later than Boyle, Lemery, Slare,

and Hawkesbee observed
that the glow is brighter
when the air is rarefied by
an air-pump, although
Lampadius showed that
it is extinguished in a
Torricellian vacuum, so
that the presence of a
trace of oxygen is neces-
sary. The dependence of
the glow on the pressure
of the gas was exhibited in
the most striking manner
by Henry and by Graham,
who made the remarkable
observation that it ceases
altogether in pure oxygen
at atmospheric pressure,
but reappears when the
pressure is reduced or an
indifferent gas added. This
may be exhibited by the
apparatus shown in Fig.
313. A stick of phosphorus
is placed in the constricted
part, a, of a tube contain-
ing oxygen confined over
mercury, the
levelling tube
being ad justed
so that the
gas is at at-
m o s p h e r i c
pressure. No
glow can be
observed in
the dark. If
the levelling

tube is now lowered so as to reduce the pressure, the phosphorus
begins to glow. In oxygen at atmospheric pressure, phosphorus
begins to glow at 25° ; the glow is very bright at 36°, and the
phosphorus then very easily inflames. The following experiment

FIG. 313.— Effect of Pressure on Phosphorescence in Oxygen.

xxxi PHOSPHORUS »i 1 7

is more convenient, as there is then no danger of the phosphorus
taking fire.

EXPT. 253.— Heat a piece of phosphorus with olrte oil in a flask 011
a water-bath. Cool the solution, and pour it into a round litre flask
fitted with a rubber stopper carrying two gas delivery tubes. Displace
the air from the flask by a current of dry oxygen. The glow ceases.
Close one tube with a piece of rubber tubing and a clip, and connect the
other with an air-pump. On reducing the pressure of the oxygen the
glow commences again.

In perfectly dry oxygen phosphorus may be distilled without change.

Graham (1829) found that the glow of phosphorus is inhibited
by the presence of ether, naphtha, or turpentine vapour. (The
action of essential oils had been observed by Boyle.) One part of
turpentine vapour in 4444 parts of air was sufficient. Later,
observers found that many essential oils, camphor, naphthalene,
carbon disulphide, and especially iodobenzene, had the same effect.

Schonbein (1848) considered that the glow is intimately related
to the formation of ozone (p. 320), since (1) essential oils which destroy
or dissolve ozone inhibit the luminosity ; (2) at low temperatures
no ozone is formed and phosphorus does not glow ; (3) at 25° the
production of ozone is a maximum, and the glow is brightest. The
exact relation between the glow and the production of ozone is not
yet settled.

The reaction occurs between phosphorus vapour and oxygen, since
it is brighter at lower pressures, and an indifferent gas (N2 or H2),
when passed over phosphorus, glows when mixed with oxygen. Some
chemists think oxygen atoms are formed : P4 + 6O2 = 2P2O5 + 2O ;
O2 + O = O3. Thorpe considers that part at least of the glow is due to
the oxidation of the lower oxide, P4O6, which is also formed, and this
is supported by Schenck. The latter considers that the lower oxide,
when formed, reacts with water to form phosphorus, phosphorous acid,
phosphoric acid, and solid hydrogen phosphide, P12H6 (P- 622). The
air also becomes ionised, i.e., it conducts electricity, and this is sometimes
considered to be due to an " emanation " given out by the phosphorus.
The action of essential oils in stopping the glow is supposed to be due to
the absorption of the P4OG by the double linkages in the compounds,
with formation of rings :

\C-0- P/u

This explanation cannot, however, cover all the cases.

The extinction of the glow in pure oxygen is put down to the oxida-
tion of P4O0 to phosphoric acid.

618

INORGANIC CHEMISTRY

Hydrogen phosphides. — Phosphorus forms with hydrogen four
compounds :

PH3, Trihydrogen phosphide (gaseous phosphoretted hydrogen :
phosphine), m.-pt. — 133°, b.-pt. — 85°.

P2H4, Di hydrogen phosphide (liquid phosphoretted hydrogen),
m.-pt. -10°, b.-pt. 57°.

Two varieties of solid phosphoretted hydrogen, P12H6 and P9H2.

Trihydrogen phosphide, usually known as phosphine, or phos-
phoretted hydrogen, was obtained by Gengembre. in 1783, by boiling
white phosphorus with a solution of caustic potash. Caustic
soda, lime, or baryta may also be used. The colourless gas so

FIG. 314.— Preparation of Phosphine.

obtained has a very unpleasant odour of rotten fish, and is poisonous.
It is spontaneously inflammable in air, and its production from
decaying organic matter in marshes is supposed to be responsible
for the phenomenon known as the Will-o'-the-wisp. In the above
reaction phosphine and an acid salt of hypophosphorous acid, H3P02,
e.g., sodium hypophosphite, NaH2P02, are formed :

P4 + 3NaOH + 3H2O = 3NaH2P02 + PH3.

The hypophosphite, on boiling, is partly decomposed, with libera-
tion of hydrogen, so that the gas is not pure : NaH2PO2 + 2NaOH =
2H2 + Na3PO4 (sodium phosphate) ; baryta gives a purer gas. Hydro-

fen is also evolved by the direct reaction : 2P -f- 2NaOH -f- 2H2O =
NaH2PO2 + H2.

xxxr PHOSPHORUS 619

EXPT. 254. — Pieces of white phosphorus are placed in a flask (Fig. 314)
containing a 30-40 per cent, solution of caustic soda. The air is first
swept out by a current of hydrogen or coal gas, to avoid the explosion
which would occur by the spontaneous ignition of a mixture of phos-
phine and air, and the flask is heated. Each bubble of phosphine
which escapes from the delivery tube dipping under water ignites
spontaneously with a bright flash, and a vortex -ring of white smoke,
consisting of phosphorus pentoxide, rises in the air. The experiment
is best performed in a fume-cupboard.

Small quantities of phosphine appear to be produced by heating
red phosphorus in hydrogen, or by adding bits of white phosphorus
to a mixture of zinc and dilute sulphuric acid evolving hydrogen
(nascent hydrogen). The hydrogen then burns with a green
flame. This is a delicate test for free phosphorus. The result
may be due to phosphorus vapour.

The spontaneous inflammability of the gas prepared by Gen-
gembre's method is due to the presence of traces of the liquid
hydride, P2H4 : 6P + 4NaOH + 4H20 = 4NaH2P02 + P2H4.

Davy showed that phosphine is evolved on heating phosphorous
acid, H3P03 (obtained by the action of water on phosphorus tri-
chloride) ; this gas is not spontaneously inflammable, but ignites
at 100° : 4H3P03 = 3HP03 + 3H20 + PH3.

P. Thenard, in 1845, showed that if the spontaneously inflammable
gas is passed through a tube immersed in a freezing mixture, the liquid
hydride is deposited, and the gas is no longer spontaneously inflammable.
The same result is obtained by passing the gas over recently ignited
charcoal, which absorbs the vapour of the dihydride, or by mixing the
gas with a little ether vapour. The pure gas becomes spontaneously
inflammable if mixed with a little vapour of fuming nitric acid.

A gas which is not spontaneously inflammable, but contains hydrogen
as impurity, is formed if phosphorus is heated with alcoholic potash.

Pure phosphine is prepared by heating phosphorous acid, by
warming phosphonium iodide (q.v.) with caustic potash solution :
PH4I + KOH = KI + H20 + PH3, or by the action of dilute
sulphuric acid on aluminium phosphide (prepared by heating
aluminium powder and red phosphorus). It is sparingly soluble in
water, alcohol, or ether.

The normal density of phosphine is 1-52058 gm./lit. It is decom-
posed by electric sparks, depositing red phosphorus, and increasing
in volume in the ratio 2 : 3 : — 2PH3 — 2P -f 3H2. In this way its
composition is determined. The gas is also decomposed by heating
to 440°.

If phosphine is kindled in a test-tube, it burns with deposition
of phosphorus : the heat of combustion of part of the gas decom-

620

INORGANIC CHEMI8TB Y

CHAT.

poses the rest (cf. HjS). A mixture of pure phosphine with oxygen
is not spontaneously explosive, but if the pressure is reduced, a
violent explosion occurs (Labillardiere, 1817).

Phosphine ignites spontaneously in chlorine. It combines with
many metallic chlorides. The pure gas, is completely absorbed by
a solution of bleaching powder. It precipitates phosphides from
solutions of many metallic salts (e.g., CuS04, AgNO3). These
phosphides are also formed by heating the solutions with white
phosphorus.

EXPT. 255. — Boil a few pieces of white phosphorus with a solution of
copper sulphate. Black cupric phosphide, CusP2, is formed.

Phosphonium compounds. — Although phosphine has a neutral
reaction to litmus paper, it is capable of acting as a feeble base,

JTIG. 315.— Preparation of Phosphonium Iodide.

forming phosphonium salts with halogen hydracids : PH3 -f- HX =••
PH4X, analogous to ammonium salts, NH4X. A mixture of phosphine
and dry hydrogen chloride does not react at atmospheric pressure,
but if cooled to — 35°, or compressed to 18 atm. at 15°, it deposits
white crystals of phosphonium chloride, which dissociate again
on warming or on reducing the pressure : PH3 -f HC1±^PH4C1.
Phosphonium bromide, PH4Br, is more stable, and is produced in
cubic crystals when a mixture of PH3 and HBr is led into a mode-
rately cooled flask. Phosphonium iodide, PHJ. is a fairly stable
compound, and is formed on mixing PH3 and HI at the ordinary
temperature and pressure. It dissociates at 30°, but the crystals
can be sublimed. Phosphonium iodide is most conveniently
prepared by the following process.

EXPT. 256. — One hundred parts of white phosphorus are dissolved
in an equal weight of carbon clisulphide in a tubulated retort, from which

XXXI

PHOSPHORUS

621

the ah' has been removed by a current of dry carbon dioxide. One
hundred and seventy-five parts of iodine are then added, and the carbon
disulphide is distilled off on a water-bath in a current of CO2. The neck
of the retort is then connected with a wide glass tube and receiver, and,
by means of a dropping -funnel fitted in the tubulure of the retort, 85
parts of water are dropped gradually on to the phosphorus iodide
(Fig. 315). Phosphonium iodide sublimes into the wide tube ; the retort
is gently warmed at the end of the process. Two wash-bottles
containing water are attached to the receiver, to absorb the hydriodic
acid evolved : 9P + 51 + 16H2O = 4H3PO4 -f- 5PHJ.

Phosphonium iodide is at once decomposed by water or alkalies,
evolving pure phosphine : PH4I -f- Aq. = PH3 -f HIAq.

Liquid phosphoretted hydrogen, P2H4. — This substance is
prepared by the action of warm water on calcium diphosphide :
Ca2P2 + 4H20 = 2Ca(OH)2 + P2H4.

Calcium phosphide, Ca2P2, is formed as a dark brown solid, con-
taining Ca3P2 and the pyrophosphate, Ca2P207. by passing phosphorus
vapour over fragments of quicklime strongly heated in a hard glass
tube.

Tricalcium diphosphide, Ca3P2, is obtained in a pure state by heating
calcium and phosphorus together under petroleum. It gives pure
phosphine, not spontaneously inflammable, when treated with water.

EXPT. 257. — If pieces of calcium phosphide are dropped into warm
water, gaseous phosphoretted hydrogen, PH3, containing the vapour of
the dihydride, P2H4, is evolved, and each bubble ignites spontaneously
as it breaks on the surface of the water.

Tin canisters filled with calcium phosohide, attached to wooden
floats, are sometimes used at sea for signalling. The canister is pierced
above and below and thrown overboard. The gas ignites spontaneously
and burns with a luminous flame (Holmes's signal).

In the pre-
paration of
liquid phos-
phoretted hy-
drogen, pieces
of calcium
pho sph ide
are dropped
through a
wide tube
into water

at 60° in a Woulfe's bottle (Fig. 316), the air having been previously
displaced by hydrogen. The gas is passed through a cooled tube

FIG. 316.— Preparation of Liquid Phosphoretted Hydrogen.

«L>2 INORGANIC CHEMISTRY THAI'.

to deposit moisture, and the liquid phosphoretted hydrogen then
condensed in a second tube cooled in a freezing mixture.

The empirical formula of liquid phosphoretted hydrogen is PH2 ;
the formula P2H4 is given to the substance by analogy with hydr-
azine, N2H4, although the vapour density cannot be found, since the
vapour is too unstable. The liquid also decomposes on exposure
to light ; gaseous phosphine is evolved, and red solid hydrogen
phosphide, P12H6, deposited: 15P2H4 = P12H6 + 18PH3. The
same solid is formed if the uncondensed vapours from the prepara-
tion of the liquid are passed into a large flask containing a little
fuming hydrochloric acid.

Solid hydrogen phosphides, P12H6 and P9H2. — The red solid
hydride, prepared as described above, is found, from the depression
of freezing point of white phosphorus in which it is dissolved, to be
P12H6. When heated in a vacuous tube it evolves pure phos-
phine and leaves a second red solid hydride, P9H2 : 5P12H6 =
6P9H2 + 6PH3. A third solid hydride, P5H2, is said to be formed
by the action of very dilute acetic acid on the phosphides of alkali
metals, e.g., Na2P5 (p. 793),

By digesting white, or scarlet, phosphorus with alcoholic potash and
water, a dark red solution is formed, which appears to contain potass-
ium polyphosphides, K2Pn. Phosphine is evolved, together with
hydrogen, and hypophosphite is also formed. On acidifying the solu-
tion, a reddish-yellow precipitate, formerly considered to be a sub-
oxide, P4O, but probably impure solid hydrogen phosphide mixed with
red phosphorus, is thrown down. P12H6 dissolves in alkalies to form
red solutions, and forms a dark-coloured compound with piperidine,
P12H6(C5HnN)3. It therefore behaves as a weak acid. .

Halogen compounds of phosphorus. — Phosphorus forms two series
of halogen compounds, in which it is tervalent and quinquevalent,
respectively : PX3 and PX5. These are obtained by the direct
combination of phosphorus and the halogen, PX3 or PX5 being
formed according as the former or the latter is in excess. The
physical properties of these compounds (iodine also forms P^^,
analogous to P2H4) are given below :

PF3, colourless gas, b.-pt. — 95°, m.-pt. — 160°.
PF5, colourless gas, b.-pt. — 755°, m.-pt. — 83°.
PC13, colourless liquid, b.-pt. 76°, m.-pt. — 112°.

PC15, white, crystalline solid, sublimes: m.-pt. 148°, b.-pt. 162° in
sealed tube

PBr3, colourless liquid, b.-pt. 170-8°, m.-pt. — 41-5°.
PBr5, orange-yellow, crystalline solid, decomposes on heating.
P2I4, orange-red crystals, m.-pt. 1 10°.
PI3, dark red crystals, m.-pt. 55°.

XXXi

PHOSPHOKUS

The compounds PBr7, PBr2F3, PCl3Br2, PCl3Br2-Br2, etc., are also

known.

Phosphorus trifluoride, PF3, is obtained by the action of arsenic

trifluoride (q.v.) on phosphorus trichloride : AsF3 + PC13 = AsCl3 -f-

PF3 ; by warming

phosphorus tri -

bromide with zinc

fluoride : 3ZnF2 +

2PBr3 = 2PF3" +

3ZnBr2 ; or by

heating copper

phosphide with

lead fluoride. The

gas has no action

on glass in the

cold ; it is hydro- FIG> 317 —Preparation of Phosphorus Trichloride.

lysed by wrater :

PF3 + 3H2O = H3PO3 + 3HF. The pentafluoride, PFft, is formed

when phosphorus burns in fluorine ; when arsenic trifluoride is

added to phosphorus pentachloride in a freezing mixture :

3PC1- + 5AsF3 = 3PF5 -f 5AsCl3 ; or when phosphorus fluorbromide

(obtained by cooling a mixture of bromine and PF3 to — 20°) is warmed

to 15° : 5PF3Br2 = 3PF5 + 2PBr5. The density of the gas is normal,

corresponding with the formula PF5, and this confirms the quinque-
valency of phosphorus. The gas does not attack
glass, fumes in the air, forming POF3, and combines
directly with ammonia gas, forming a solid :
2PF5,5NH3.

Phosphorus trichloride, PC13 (Gay-Lussac and
Thenard, 1808), is formed (along with some
pentachloride) when phosphorus burns spon-
taneously in chlorine. It is made by passing
a stream of dry chlorine over white or red
phosphorus in a retort, and condensing the
product in a dry cooled receiver (Fig. 317).
It is purified by standing over white phosphorus,
and redistilling. The pure liquid is colourless,
and may be preserved in sealed flasks. The
vapour density is normal. It fumes strongly in
moist air : PC13 + 3H20 = 3HC1 + H3P03
(phosphorous acid).

Phosphorus pentachloride, PC15 (Davy, 1810 ; Dulong, 1816), is pre-
pared [ExPT. 258] by passing dry chlorine through a large cooled
flask, into which the trichloride is allowed to drop from a tap-funnel
(Fig. 318). It is a greenish-white solid, which sublimes at the

FIG. 318.— Prepara-
tion of Phosphorus
Pentachloride.

624 INORGANIC CHEMISTRY CHAP.

ordinary pressure below 100° without previous fusion, the vapour-
being dissociated into trichloride and chlorine (p. 153) :
PC15 ;=± PC13 + C12. Above 300° the dissociation is practically
complete. If heated under pressure, it melts at 148°. Although
the compound is not, as was formerly supposed, a molecular com-
pound PC13,C12, two atoms of chlorine are very reactive, and many
metals (Zn, Cd. and even Au and Pt) are converted into chlorides,
PC18 being left : PC15 -f Zn = ZnCl2 + PCla.

Phosphorus tri- and penta -chlorides are violently hydrolysed
by water, the reactions being irreversible (p. 450). The trichloride
is completely freed from halogen, and phosphorous acid, H3PO3,
is formed (with a small quantity of water, a trace of POC1 is said
to be formed) : PC13 -f- 3H20 = H3P03 + 3H01. In the case of
the pentachloride the reaction proceeds in two stages. With a little
water, liquid phosphorus oxychloride, or phosphoryl chloride, POC13, is
produced, which is further hydrolysed by excess of water with
formation of orthophosphoric acid, H3P04 :

PCL + H2O = POC13 -f 2HC1.
POC13 + 3H20'= H3P04 + 3HC1.

If excess of water is added to the pentachloride, phosphoric acid
is produced, although the oxychloride is probably formed as an
intermediate product : PC15 + 4H2O = 5HC1 + H3P04.

Inorganic oxy -acids, organic acids (containing the carboxyl group,
— CO -OH, p. 518), and alcohols (hydroxides of hydrocarbon radicals,
e.g.,, methyl ' alcohol, CH3'OH)3 containing the hydroxyl group,
OH, react with phosphorus tri- or penta- chloride, the hydroxyl
group being eliminated and substituted by an atom of chlorine. This
reaction is frequently applied in organic chemistry to the detection
of hydroxyl groups in compounds :

3C2H5-OH + PC13 = 3C2H5C1 -f H3P03.

Ethyl alcohol Ethyl chloride

CH3-CO-OH + PC15 - CH3-CO-C1 -f POC13 + HC1.

Acetic acid Aeetyl chloride

Acetone, (CH3)2CO, which does not contain a hydroxyl group, reacts
with phosphorus pentachloride, but the oxygen atom alone is replaced
by two atoms of chlorine :

(CH3)2CO + PC15 = (CH3)2CC12 + POC13,
Acetone Dichloropropane

Sulphur trioxide reacts violently with phosphorus trichloride : SO3 +
PC13 = SO2 -f- POC13. Phosphorus pentachloride reacts with dry
ammonia, forming ammonium chloride and chlorophosphamide,
PC13(NH2)2. The latter is converted by water into phosphamide,
PONH-NH2, a white powder insoluble in water, dilute acids, and alka-
lies. If phosphamide is heated in absence of air, phospham, (PN2H)3;,
remains as a white powder, which is only very slowly oxidised on heating

xxxi PHOSPHORUS 625

to redness in air. It is decomposed with incandescence by fused alkalies,
ammonia and a phosphate being formed.

At 175-200°, ammonia and phosphorus pentachloride form a mixture
of six phosphonitrile chlorides : (PNC12)3, (PNC12)4, (PNC12)5, (PNCl2)fl,
(PNC12)7, and (PNC12)^, which are very stable. The main product is
(PNC12)3, b.-pt. 256°, m.-pt. 114°. Ethereal solutions of these com-
pounds, when shaken with water, form metaphosphimic acids ; stable
baits, e.g., P3N3O6H3(NH4)3 + H2O, are known.

White phosphorus explodes in contact with bromine ; liquid
bromine dropped on red phosphorus in a cooled flask reacts with
evolution of light, and the tribromide, PBr3, distils over. By adding
bromine to this, the solid pentabromide is formed. The latter is
also formed by the action of bromine and iodine on the trichloride ;
iodine chloride is also formed : the reaction, 2PC13 -f- 3Br2 ±1:
2PBr3 -j- 3C12, takes place to a slight extent, and the C12 is removed
by the iodine as fast as it is produced. The solid exists in two
forms : a red variety obtained by rapidly cooling the vapour, and
a yellow stable variety obtained on slow cooling. The vapour
is dissociated : PBr5 dr PBr3 -f Br2.

White phosphorus inflames in contact with iodine ; if solutions
of iodine and phosphorus in carbon disulphide are mixed, the
di-iodide and tri-iodide, P2T4 and PI3, are obtained on evaporation.

A dichloride, P2C14, corresponding with P2I4, is said to be formed, as
an oily, fuming liquid, by the action of the silent discharge on a mixture
of PC13 vapour and hydrogen.

Sulphides of phosphorus. — Yellow phosphorus and sulphur form
spontaneously inflammable solid solutions when fused together,
but if a mixture of red phosphorus and small pieces of roll sulphur
is heated in a loosely -corked glass flask on a sand-bath, chemical
reaction commences, and then proceeds without further heating.
According to the proportions taken, the sulphides P2S5, P4S7, and
P4S3 are obtained. The pentasulphide is purified by distilling in
dry carbon dioxide ; it is a pale yellow solid melting at 275° and
boiling at 530° ; the vapour has the normal density. The sub-
stance is rapidly hydrolysed by water :

P2S5 4- 8H20 = 2H3P04 + 5H2S,

and is used in organic chemistry for replacing the hydroxyl group.
OH, in compounds by the group SH. Thus, alcohol, C2H5OH,
forms mercaptan, CgHg-SH. Tetraphosphorus trisulphide, P4S3, is
purified by crystallisation from carbon disulphide or phosphorus
trichloride, or by distillation in vacuo. It melts at 172-5°, boils
at 408°, giving the normal vapour density, and is only slowly
hydrolysed by water. P4S7 forms slightly yellow crystals from
CS2, m.-pt. 310°, b.-pt. 523°.

s s

(Wti INORGANIC CHEMISTRY CHAP.

Matches. — Common lucifer matches are made by dipping thin
strips of wood, cut by machinery and coated at one end with paraffin
wax or sulphur, into a paste of yellow phosphorus, gum, red lead,
and sometimes potassium chlorate. Bundles of splints are dipped
at once, and then dried. The heads ignite when rubbed on sand-
paper, the local heating bringing about combustion. On account
of the poisonous properties of yellow phosphorus, it is being replaced
by scarlet phosphorus (p. 614) : the sulphide, P4S3, is also used.
Safety matches are dipped into a paste of 24 parts of antimony
sulphide, 32 of potassium chlorate, 12 of potassium dichromate,
32 of red lead, and gum. The heads contain no phosphorus.
They are rubbed on a strip of paper coated with red phosphorus,
antimony sulphide, powdered glass, and gum, attached to the box.
They may also be ignited by drawing rapidly over glass or linoleum.
The wood is often impregnated with borax, so that it does not glow
after the flame is blown out.

Oxides and oxy-acids of phosphorus.— Three oxides and several
cxy-acids of phosphorus are known :—

Hypophosphorous acid, H3P02.
Phosphorus trioxide, P2O3 or Phosphorous acid, H3P03.

P40«.

Phosphorus tetroxide, P2O4. Hypophosphoric acid, H2P03

Phosphorus pentoxide, P205 or Phosphoric acids :

P4O10, the anhydride of P205 + 3H20=2HJ>04,ortho-

three phbsphoric acids. phosphoric acid ;

P205 + 2H20 - H4P207, pyro-

phosphoric acid ;
P2O5 + H20 == 2HP03. meta-
phosphoric acid, of which
polymeric modifications,
(HP03)W, are known.
Permonophosphoric acid, H3P05 ;
Perphosphoric acid, H2P2O8.

The so-called phosphorus sub-oxide, P4O, is probably a mixture of
red phosphorus and the solid hydride, P12HC (p. 622). Leverrier's oxide
(1838), obtained as a red powder by allowing sticks of phosphorus
partly covered with phosphorus trichloride to stand in a flask of air, is
probably red phosphorus.

The burning of phosphorus.— When phosphorus is burnt in a free
supply of air, phosphorus pentoxide, P205, first observed by Boyle and
called " flowers of phosphorus," is formed. During the later stages
of the combustion of phosphorus in a limited supply of air, phos-
phorus trioxide, P2O3, is formed. The phosphorus is extinguished
before all the oxygen is removed, and a portion of the phosphorus
is converted into red phosphorus

PHOSPHORUS fi27

EXPT. 259. — Dry the air inside a tall bell- jar by means of a capsule
of sulphuric acid standing on a ground glass plate supporting the jar.
After a few hours remove the capsule, and replace it by a small porcelain
crucible-lid supported on a cork, in which a tit of phosphorus is placed.
The phosphorus is ignited by touching with a hot wire as it is placed
under the jar. Notice the bright flame, and the formation of a snow-
white powder (P2Ofi). which rapidly settles. After a time, the flame
becomes larger, greenish, and flickering : P2O3 is then formed. Finally
it goes out.

Phosphorus pentoxide, P205. — This oxide is always prepared by
the combustion of phosphorus in air
or oxygen. On a large scale, the ap-
paratus shown in Fig. 319 is used.

The sheet -iron cylinder is provided
with an opening at the side, through
which a copper spoon containing
phosphorus is introduced. The phos-
phorus is ignited, and the pentoxide
produced settles out, and falls into the
dry bottle below. More phosphorus is
added from time to time by drawing
out the spoon, and the lid is raised to
renew the air.

The voluminous powder so ob-
tained, when heated to 440°, be-
comes more compact and less volatile.
If distilled in dry carbon dioxide, the
pentoxide forms crystals subliming at 250°. The compact variety
melts under pressure at a red heat, forming a vitreous mass.

Commercial phosphorus pentoxide contains some trioxide, P2OS, and
metaphosphoric acid. It may be purified by volatilising in a current

FIG. 319. — Preparation of
Phosphorus Pentoxide.

Aspirator

FIG. 320.— Preparation of Pure P200.

of oxygen in a hard glass tube, passing the gas over heated platinised
pumice, and condensing in a cooled receiver. The following' method is
more convenient :

Dry red phosphorus is sealed up in a hard glass tube with a capillary
tip, A (Fig. 320), placed in the hard glass tube, B. A slow stream of dry
air is passed through B, and the part under the phosphorus heated till a

s s 2

«2S INORGANIC CHEMISTRY CHAP.

small flame appears at the capillary. The narrow part of the tube JB,
containing a spiral of platinum wire, is heated to redness. The pent-
oxide is collected in the tube C, which is plugged with glass wool, and is
withdrawn when filled.

The vapour density of phosphoric oxide at 1400° is slightly
higher than corresponds with the formula P4010, but since the
molecular weight in the solid state is not known, the simpler formula
P2O5 is generally used.

Phosphorus pentoxide exhibits a strong phosphorescence after
illumination ; the effect is more marked at low temperatures. Its
most characteristic chemical property is its powerful affinity for
water. The solid rapidly becomes moist and sticky on exposure
to air, metaphosphoric acid, HP03, being formed, and it withdraws
the last traces of moisture from gases dried with calcium chloride,
caustic potash, or sulphuric acid. When thrown into water,
phosphorus pentoxide dissolves with a hissing noise and the evolu-
tion of much heat (cf. S03, p. 498). Phosphorus pentoxide with-
draws the elements of water from many acids, and other substances
containing hydrogen and oxygen, forming anhydrides (e.g., SO3
from H2S04,* N2O5 from HN03, C1207 from HC104). It may, in
these reactions, continuously remove traces of water already pro-
duced by dissociation of the acids : this is undoubtedly the case
with nitric and sulphuric acids : 2HN03 ~ N205 + H2O.

The phosphoric acids. — Phosphorus pentoxide is the anhydride
of the phosphoric acids, three of which are known :

P2O5 + H2O = 2HP03, metaphosphoric acid.
P205 + 2H2O = H4P2O7, pyrophosphoric acid.
P2O5 + 3H20 = 2H3P04, orthophosphoric acid.

These may also be regarded as products of dehydration
of a hypothetical acid, >(OH)5 : P(OH)5 - H20 = H3P04 ;
2H3P04 - H20 - H4P207 ; H4P2O? - H20 - 2HP03.

Metaphosphoric acid, HP03, is formed as a viscous mass when
the anhydride is exposed to moist air, or moistened with cold water :
P205 4- H20 =- 2HP03. If a solution in water is boiled, meta-
phosphoric acid is converted into orthophosphoric acid :

HP03 + H20 = H3P04.

This change occurs slowly on standing in the cold, pyrophosphoric
acid being formed as an intermediate product. The natural mineral
phosphates, and bone-ash (p. 609), are salts of orthophosphoric
acid, and this was the first phosphoric acid to be prepared. The
fertiliser guano, consisting of the excreta of sea birds, is rich in
phosphates, and also in combined nitrogen. Another source of
phosphates, used for fertilisers, is the basic slag, Ca4P209, of steel
furnaces (p. 981).

XXXI

PHOSPHORUS

629

When disodium orthophosphate, Na2HPO4. is heated to redness,
ifc forms the sodium salt of pyrophosphoric acid :
2Na2HP04 - H20 = Na4P2O7.

Orthophosphorie acid, H3P04. — This acid is prepared technically
by digesting 100 parts of bone-ash with a mixture of 96 parts of
concentrated sulphuric acid and 1000 parts of water for several
hours: Ca3(P04)2 + 3H2S04 = 3CaS04 + 2H3PO4. The calcium
sulphate is filtered off, and the phosphoric acid evaporated to a
specific gravity of 1 -7 (85 per cent. H3P04). The product is impure,
containing acid calcium phosphate, CaH2(P04)2, which may be
removed by adding concentrated sulphuric acid, filtering, evaporat-
ing, and driving oft' the volatile sulphuric acid by ignition. The
fused mass on cooling solidifies to a glass of metaphosphoric acid,
(HP03)«, usually called glacial phosphoric acid. It contains a little
magnesium phosphate.

Pure orthophosphoric acid is obtained by the oxidation of phos-
phorus with nitric acid. Oxides of nitrogen are evolved.

EXPT. 260. — Five gm. of red phosphorus are heated with 50 c.c. of
concentrated nitric acid in a flask provided with a reflux condenser
fitted with a ground glass stopper (Fig. 321).
Red fumes of oxides of nitrogen are evolved,
so that the experiment is performed in a fume-
cupboard. When the phosphorus has dissolved,
the liquid is evaporated in a platinum dish on a
sand-bath, a little concentrated nitric acid is
added to oxidise any phosphorous acid, and the
liquid is then evaporated and heated to drive
off nitric acid. The glassy residue is dissolved
in water, and evaporated in a platinum dish
until the temperature rises to 150°. On cooling
the syrupy liquid, hard rhombic crystals of
orthophosphoric acid, H3PO4, are slowly de-
posited. If the phosphorus used contains
arsenic, the solution of phosphoric acid is
treated with sulphur dioxide to reduce arsenic
acid to arsenious acid, the excess of sulphur
dioxide is expelled by boiling, and the arsenic
precipitated as sulphide by H2S. The filtered
solution is evaporated. If yellow phosphorus is used, nitric acid
of sp. gr. 1-2 is employed, to avoid explosions, and a trace of iodine
may be added as a catalyst. If the temperature is carried beyond
150° in the evaporation, some metaphosphoric acid is formed, which
retards crystallisation.

The crystals of orthophosphoric acid melt at 38-6°, and are very

FIG. 321 .—Preparation of
Phosphoric Acid from
Phosphorus.

630 INORGANIC CHEMISTRY CHAP.

soluble in water. Two crystalline hydrates, 2H3P04,H2O and
10H3P04,H20, are known. The aqueous solution has a strong,
purely acid, taste and no smell, and has been used for making
'• lemonade."

The orthophosphates. — Orthophosphoric acid is tribasic, and forms
three series of salts :

Prirdary orthophosphates, e.g.. sodium dihydrogen phosphate,

NaH2P04.
Secondary orthophosphates, e.g., disodium hydrogen phosphate,

Na2HP04.
Tertiary orthophosphates, e.g., trisodium phosphate, Na3P04.

Or^ophosphates are usually called simply " phosphates."
Ordinary sodium phosphate is the secondary salt, Na2HP04,12H20.

The alkali phosphates (except lithium phosphate, Li3PO4) are
soluble in water. The tertiary phosphates of the remaining metals
are insoluble in water, but dissolve in dilute mineral acids :
Ca3(P04)2 + 6HC1 — 3Ca012 + 2H3PO4. If the acid solutions are
neutralised, the phosphates are reprecipitated :

3CaCl2 -f 2H3P04 + GNaOH =-Ca3(P04)a + GNaCl + 6H20.

Aluminium and ferric phosphates are insoluble, chromium phos-
phate is sparingly soluble, and the remaining phosphates are soluble,
in acetic acid. If to a solution of a phosphate in acetic acid ferric
chloride is added, the phosphoric acid is therefore precipitated as
ferric phosphate, and is removed from the solution :

3Ca(C2H302)2 -f 2H3P04 + 2FeCl3 =
2FePO4 (pp.) + 3CaCl2 + 6C2H4O2.

An excess of ferric chloride then forms a blood-red solution of ferric
acetate, Fe(C2H3O2)3, but on boiling the whole of the iron is pre-
cipitated as basic ferric acetate ; the filtrate contains the other
metals (except Al and Or).

In qualitative analysis, if a solution contains phosphoric acid, this
must be removed before adding ammonia to precipitate Group III
(Fe, Al, Cr) ; otherwise the phosphates of the remaining groups (except
the alkali-metals) would also be precipitated. The solution from the
sulphuretted hydrogen precipitation is boiled with nitric acid to oxidise
ferrous salts, nearly neutralised with sodium carbonate, and then a
mixture of sodium acetate and acetic acid added. A1PO4, FePO4, and
CrPO4 are precipitated. The filtrate is then treated with ferric chloride
until the deep red colour of ferric acetate appears. It is boiled, and the
filtrate is free from phosphates.

The primary soluble phosphates in solution are acid to litmus ;

xxxi PHOSPHORUS 631

tertiary phosphates are alkaline ; whilst secondary phosphates are
faintly alkaline — practically neutral :

H2P04" =r±HP04" + H\

PO /" + H20 = HP04" + OH'.

(HPCY 4- H' + OH' - - H2P04' + OH'.)

The first two hydrogen atoms of orthophosphoric acid are easily
ionised in solution : the third is split off only with difficulty, and in
presence of an excess of base :

H3P04 ^± H' + H2P04' ^± 2H' -f HP04" ^ 3H' + P04'/7.

On titration with litmus, phosphoric acid therefore behaves as a
dibasic acid. Methyl-orange, however, changes colour at the
stage NaH2P04 ; phenol phthalein at the stage Na2HP04 ; the
changes occur sharply at 55°.

Solutions of phosphates, when treated with excess of nitric acid
and a solution of ammonium molybdate (p. 957), slowly deposit
in the cold a canary-yellow precipitate of ammonium phosphomolyb-
date, readily soluble in ammonia.

Pyro- and met a -phosphates also give this reaction ; they are first of all
converted by the reagent into orthophosphoric acid. Arsenic acid,
H3AsO4, gives a similar precipitate, but only on heating. The precipita-
tion of phosphoric acid also occurs much more rapidly on heating.

Ordinary sodium phosphate, Na2HPO4,12H20, is prepared by
neutralising phosphoric acid with caustic soda or sodium carbonate
(the end-point should be faintly alkaline), and evaporating.
It forms efflorescent crystals, m.-pt. 35°, readily soluble in
water.

If a solution of phosphoric acid is divided into three parts, the
equivalent quantities of caustic soda and ammonia, respectively,
added to two to form NanP04 and (NH4)3PO4, and all three solutions
mixed and evaporated, crystals of microcosmic salt, or sodium
ammonium hydrogen phosphate, NaNH4HPO4,4H2O, are formed.
This salt may also be prepared by dissolving 6 gm. of
ammonium chloride and 36 gm. of ordinary sodium phosphate
in a little hot water, filtering off the sodium chloride, and
crystallising.

The primary, or acid sodium phosphate, NaH2P04,H20, is prepared
by adding phosphoric acid to a solution of the ordinary phosphate,
until the solution no longer precipitates barium chloride, and
evaporating; it is dimorphous. Trisodium phosphate is prepared
by dissolving the calculated amounts of sodium phosphate
and caustic soda in hot water, and evaporating ; crystals of
Na3PO4,12H2O separate These are not efflorescent or deliquescent.
This salt is used, under the name of " tripsa," for softening boiler-
water. The calcium bicarbonate is precipitated as carbonate- by

632 INORGANIC CHEMISTRY CHAP.

the alkali formed by hydrolysis, arid calcium and magnesium
chlorides and sulphates are precipitated as phosphates.

Pyrophosphorie acid, H4P207.— This acid is formed (with a little
metaphosphoric acid) when orthophosphoric acid is heated to 213° ;
condensation occurs, and a molecule of water is eliminated from
2 molecules of orthophosphoric acid : 2H3PO4 = H4P2O7 -f H20.
If ordinary sodium phosphate is heated to dull redness, it also loses
a molecule of water and forms sodium pyrophosphate (Clark, 1827) :
2Na2HP04 =•- Na4P207 -f- H20. Whereas the orthophosphate gives
a yellow precipitate of silver orthophosphate, Ag3PO4, with silver
nitrate, the residue after ignition, when dissolved in water, gives
with that reagent a white crystalline precipitate of silver pyro-
phosphate, Ag4P2O7. If lead nitrate solution is added to a solution
of sodium pyrophosphate, a white precipitate of lead pyrophosphate,
Pb2P207. is thrown down : this, when suspended in water and
treated with sulphuretted hydrogen, gives a black precipitate of
lead sulphide, and a solution of pyrophosphoric acid :

Pb2P207 + 2H2S = 2PbS + H4P207.

The solution may be evaporated in vacuo, and on cooling to
- 10° for some time yields white granular crystals of pure pyro-
phosphoric acid, H4P207. m.-pt. 61°.

If a solution of orthophosphoric acid or an orthophosphate,
mixed with ammonium chloride, is made alkaline with ammonia,
and a solution of a magnesium salt (MgCl2 or MgS04) added,
a white crystalline precipitate of magnesium ammonium phosphate,
MgNH4P04,6H20, is formed. In dilute solutions, this is deposited
slowly ; the precipitation is accelerated by adding excess of
ammonia and scratching the sides of the beaker with a glass rod.
When heated to redness, the precipitate loses ammonia and wrater
and forms a white insoluble powder of magnesium pyrophosphate :
Mg2P2O7. These reactions are utilised in the detection and estima-
tion of orthophosphoric acid or magnesium. With manganese
salts, MnNH4PO4,6H2O and Mn2P2O7 are formed.

If a solution of pyrophosphoric acid is kept for some time, or is
boiled, orthophosphoric acid is formed : H4P207 -f H2O =2H3P04.
The salts, however, are very stable in solution.

Pyrophosphoric acid contains four hydrogen atoms, and is
tetrabasic. Only two series of salts are, however, known, viz., the.
normal salts, M4P2O7, and the diacid salts, M2H2P2O7. Examples
are : Na4P207?10H20 (monoclinic) ; Na.2H2P2O7,6H2O (hexagonal) ;
Ca2P207,4H20 (amorphous, insoluble) ; Ag2H2P2O7 (soluble). Com-
plex ions containing metals (Zn, Pb, Ag, etc.) are formed by dis-
solving the insoluble pyrophosphates in sodium pyrophosphate
solution.

Metaphosphoric acid, HP03. — This acid is formed as a glassy

xxxi PHOSPHORUS <>33

residue when either ortho- or pyro -phosphoric acid is heated to
redness : H3P04 = HPO3 + H2O (Graham, 1833). By prolonged
heating, some phosphoric anhydride appears to be produced, and
the hard glass formed on cooling crackles when thrown into
water (Berzelius). The water content of the residue depends 011
the duration of heating ; pyrophosphoric acid is formed as an inter-
mediate product. At a white heat, the acid volatilises. If the
glass is dissolved in water, the freezing-point depression shows
that the acid is polymerised, (HPO3)n,- whereas the solution of
the acid prepared from the insoluble lead salt and hydrogen
sulphide (cf. pyrophosphoric acid) has the simple molecular
weight, HPO3.

Sodium metaphosphate is formed as a clear glass when microcosmic
salt, acid sodium orthophosphate, or acid sodium pvrophosphate is
heated to redness : NaNH4HP04 = NaPO3 + NH, -f H20 ;
NaH2P04 - NaP03 + H2O.

If a little microcosmic salt is heated on a loop of platinum wire, a
fused bead of NaPO3 remains, which dissolves many metallic oxides with
the formation of orthophosphates possessing characteristic colours
(" microcosmic bead ") : CoO -f NaPO3 = CoNaPO4 (blue).

A nearly neutralised solution of a metaphosphate gives a white
gelatinous precipitate of silver metaphosphate, AgP03, with silver
nitrate.

Metaphosphoric acid glass, in solution, appears to have a high
molecular weight, (HPO3)n, and behaves in many ways as a colloid.
Unlike the other phosphoric acids, it at once coagulates albumin
(white of egg), and gives white precipitates with calcium and barium
chlorides.

The metaphosphates are much more numerous than the simple
formula of the acid HPO3 would indicate, and both polymeric and meta-
meric varieties (p. 496) appear to exist. They were investigated by
Fleitmann and Hermeberg (1848), who regarded them as derived from
polymerised acids, (HPO3)H, where n = 1, 2, 3, 4, 5, and 6. Later
investigations of Tammann (1890) showed that metamerism was also
exhibited. A table of these compounds is given in Abegg's " Hand-
buch," vol. III., [3], p. 448. Sodium metaphosphate prepared from
microcosmic salt appears to be (NaPO3)i{ ; its solution is unstable. Holt
and Myers (1911), by the freezing-point method, differentiated four
varieties of metaphosphoric acid: (1) HPO3, from the lead salt and
H2S ; (2) the " crackling " acid ; (3) the non- deliquescent glass pre-
pared by heating (2) to redness for twenty -four hours, (HPO3)2 ; (4) the
deliquescent glass obtained by heating the commercial acid for a short
time, (HPO3)3.

634 INORGANIC CHEMISTRY CHAP.

The relations between the different phosphoric acids is summarised
in the following diagram :

230°

Ut I

rtv- xi3iru4 " pyro-

-tlii'sU- nwuww

i — - — > ?neto- J

boil solution

2*" superheated steam >
Pb2P207

+ |Pb(N03)2 I

I >

* Na3P04

Na4P207 \

Na2I

X+NaOH
IPO

l i

V heat

+ HoPO4 \
^ NaHoPO,

Pb(P03)2
Pb(N03),

If phosphorus pentoxide is added to 30 per cent, hydrogen peroxide,
cooled in ice, monoperphosphoric acid, H3PO5, or PO(OH)2-OOH,
analogous to Caro's acid (p. 520), is formed. Pyrophosphoric acid gives
a small quantity of a crystalline perphosphoric acid, H4P2O8, analogous
to H2S2O8.

Basicity of acids. — Until Graham's researches (1833), the three
varieties of phosphoric acid were regarded as isomeric, and, since
they were considered to enter into the salts as anhydrous oxides,
were formulated as a P205, 6 P205, and c P2O5. Graham found
that the phosphates, with the exception of the metaphosphates,
tertiary sodium phosphate, and sodium pyrophosphate, contain
hydrogen, which he regarded as present in the form of combined
water. He therefore supposed that the free acids are also com-
pounds of the anhydride with varying definite proportions of water :
meta- P205,H2O ;' pyro- P205,2H2O ; ortho- P2O5,3H20. Liebig
(1838) then pointed out that the facts could be even more simply
explained on Davy's hydrogen theory of acids, but it was then
necessary to assume that the hydrogen in orthophosphoric acid, for
instance, could be replaced in three stages, or, as Liebig expressed
it, this compound is a tribasic acid :

Graham. Liebig.

Orthophosphoric acid P2O5,3H2O H3PO4

Acid sodium phosphate P2O5,Na2O,2H2O H2NaPO4

Ordinary sodium phosphate ... P2O5,2Na2O,H2O HNa2PO4
Trisodium phosphate P2O5,3Na2O Na3PO4

Phosphorus oxychloride, POC13.— When phosphorus pentachloride
is treated with small quantities of water until the solid is com-
pletely liquefied, a colourless fuming liquid, b.-pt. 107°, m.-pt.

xxxi PHOSPHORUS 635

- 1-5°, is formed which has the composition POC13 and is known
as phosphorus oxychloride : PC15 -f H2O = POC13 -f 2HC1. It is
also formed by the direct oxidation of phosphorus trichloride by
ozone, or by the gradual addition of 32 gm. of powdered potassium
chlorate to" 100 gm. of phosphorus trichloride, and then distilling :
3PC13 + KC103 = 3POC13 + KC1. Phosphorus pentachloride and
pent oxide combine to form the oxychloride when heated in a sealed
tube: P205 -f 3PC15 = 5POC13.

The formation of the oxychloride by the action of phosphorus
pentachloride on compounds containing hydroxyl groups has
already been described (p. 624) ; the action on oxalic and boric
acids is interesting in this connection, since, in the first case, the
by-products are gaseous, and are evolved, leaving the phosphorus
oxychloride, and, in the second case, the by-product is non-volatile,
so that the oxychloride may be distilled off :

C204H2 -f PC15 = POC13 -f C02 + CO + 2HTJ1.
2H3B03 -f 3PC15 = B2O3 + 3POC13 + 6HC1.

The oxychloride is readily hydrolysed by excess of water, forming
orthophosphoric acid : POC13 -f 3H20 = H3PO4 + 3HC1.

Phosphorus oxy bromide, POBr3 (solid ; b.-pt. 190°), is similarly
prepared from the pentabromide : the oxyfluoride, POF3 (b.-pt. — 40°),
is formed by the action of dry HF on P2O5, by the action of zinc fluoride
on POC13, or by the explosion of PF3 and oxygen by a powerful induction
spark ; it does not attack glass.

Constitution of phosphoric acids. — From its method of pre-
paration, phosphorus oxychloride is ascribed the formula
O - P(C13) :

(unstable)

It contains the tervalent radical phosphoryl, 0:P^— . It is more
stable than the pentachloride, which may be regarded as containing
the radical C12:P^~, since the pentachloride is decomposed on

heating, whilst the oxychloride volatilises unchanged.

Since orthophosphoric acid is produced by the action of water
on phosphorus oxychloride (or phosphoryl chloride, as it may be
called), the latcer may be regarded as the chloride of orthophosphoric
acid, just as sulphuryl chloride is the chloride of sulphuric acid
(p. 514). Phosphorus pentachloride is the chloride of a hypothetical
acid, P(OH)5, which, if it existed, would be the true or^ophosphoric

636 INORGANIC CHEMISTRY CHAP.

acid. Orthophosphoric acid, therefore, contains the radical phos-
phoryl :

/jCl"HJ -OH /OH

0=B(— JCI HJ-OH -> 0=P^-OH -f 3HC1;
NCI Hi -OH \OH

its formula may be written 0:P(OH)3. The fact that the third
atom of hydrogen is only removed with difficulty by bases is no
proof that all three hydroxyl groups cannot be similarly attached
to the phosphorus atom, since sulphuric acid, which is always
written S02(OH)2, is only slightly ionised in the second stage, except
at very great dilutions. In reactions where ions are not concerned,
all three hydroxyl groups of orthophosphoric acid may be readily
removed, as in the formation of the ethyl ester, 0:P(OC2H5)3.
Although the two possible isomers of phosphorus oxychloride,

O— PC13 and CIO— PC12, are not known, the corresponding phenyl
compounds, O:P(C6H5)3 (m.-pt. 153-5°), and C6H5O-P(C6H5)2 (b.-pt.
265°/62 mm.), have been prepared.

From the formula O:P(OH)3 for orthophosphoric acid, those of
pyrophosphoric and metaphosphoric acids, and (though with less
probability) that of phosphoric anhydride may be inferred :

/!OH"H|0X OH OH

\ II

(1) 0=P- OH HO P=0 -> 0=P— 0-P=0 +H20,

>H HO/

°H OH

PO(OH)2

or O<( -f H2O pyrophosphoric acid.

PO(OH)2

iOH

(2) 0=P— OjH

H20, or 0:PO(OH)+H20,

metaphosphoric acid.

Oo -> V__0— P/ ' + H20,
(\// N>.Q

phosphorus pentoxide.

Another suggested formula for the anhydride, which is derived from
that of pyrophosphoric acid by repeated elimination of water, is :

PO— O— PO

/ \ / \

O - O O O

\ / \ /

PO— O— PO

This corresponds with the molecular formula P4O10.

xxxi PHOSPHORUS 637

The chloride of pyrophosphoric acid, pyrophosphoryl chloride,

P2O3C14, or O<^ , is formed by oxidising phosphorus trichloride

\POC12

at a low temperature with nitrogen tetroxide, N2O4, and distilling.
Nitrosyl chloride, phosphorus pentoxide, and phosphoryl chloride are
also formed in this reaction. Pyrophosphoryl chloride is a colourless
fuming liquid, b.-pt. 210-215°, hydrolysed by water to or£/zophosphoric
acid : P2O3C14 + 5H2O == 2H3PO4 + 4HC1. By distillation under
reduced pressure, it gives metaphosphoryl chloride, PO2C1, a syrupy
liquid.

Thiophosphoric acids. — The compounds (cf. thiosulphuric acid, p. 520)
monothiophosphoric acid, H3P(SO3) ; dithiophosphoric acid, H3P(S2O2) ;
and trithiophosphoric acid, H3P(S3O), are formed as sodium salts by
adding phosphorus pentasulphide to caustic soda, and precipitating by
alcohol. At 20° the trithiophosphate, at 50° the dithiophosphate,
Na3PS2O2,llH2O, and at 90° the monothiophosphate, Na3PSO3,12H2O,
are formed. These precipitate barium ; barium and strontium ; and
calcium, barium, and strontium salts, respectively. Thiophosphoryl
chloride, PSC13, is a colourless fuming liquid, b.-pt. 125°, obtained by
heating P2S5 and PC15 : P2S, + 3PC15 = 5PSC13. It is hydrolysed by
water : PSC13 + 4H2O = H2S + 3HC1 + H3PO4.

Magnesium ammonium thiophosphates are sparingly soluble in dilute
ammonia. Dithiophosphates give a green colour with manganese and
cobalt salts ; cobalt monothiophosphate is intensely blue, and the nickel
salt bright green.

Phosphorus trioxide, P^Og. — The formation of a lower oxide of
phosphorus, usually assumed to be P203, by the slow oxidation of
phosphorus in air, or its combustion in a limited supply of air
(p. 626), was noticed by Sage (1777), but the substance was first
obtained in a pure state by Thorpe and Tutton in 1890. Phos-
phorus is burnt in a limited supply of air, and the product
condensed by cooling.

Sticks of phosphorus 1^ in. long were placed in the hard glass tube, a,
(Fig. 322), connected with the Liebig's condenser, 6, 2 ft. in length, the
inner tube of which was one inch in diameter. A plug of glass wool in
this at the end furthest from the phosphorus served to filter out the solid
pentoxide formed, whilst the trioxide was kept in the state of vapour
by circulating water at 60° in the condenser. The condenser communi-
cated with a U-tube, c, having a small bottle at the lower part, which
was immersed in pounded ice, and this was connected through a wash-
bottle, /, containing sulphuric acid with a water -pump for aspirating air
through the apparatus. The phosphorus was ignited, and a slow
current of air drawn through. The reaction was stopped when four-

038

INORGANIC CHEMISTRY

CHAP.

fifths of the phosphorus was burnt. The trioxide condensed in the
U-tube ; on warming the latter, it collected as a liquid in the bottle.

The trioxide is also formed by the action of phosphorus trichloride
on phosphorous acid.

Phosphorus trioxide is a white, waxy, crystalline solid, m.-pt. 22-5°,
b.-pt. 173-1°. The vapour density and the depression of freezing
point of benzene correspond with the formula P406 (c/. As406,
Sb406). Unless quite pure, the trioxide slowly turns red in sunlight
from separation of phosphorus. It is very poisonous, and has an
unpleasant odour of garlic. Phosphorus trioxide oxidises in air or
oxygen at the ordinary temperature, forming the pentoxide ; at
70° it inflames in air. Under reduced pressure it glows in air.
ionising it, but not forming any ozone (c/. phosphorus vapour).
If heated in oxygen it burns ; in chlorine it inflames spontaneously,
forming POC13, and the chloride of metaphosphorous acid, P02C1,

it*

FIG. 322.— Preparation of Phosphorus Trioxide.

or possibly a mixture of P203C14 (p. 637) and P7015C15. In cold
water, phosphorus trioxide dissolves slowly (cf. P205), forming
phosphorous acid, H3P03, of which it is the anhydride. Hot water
brings about explosive decomposition, with formation of phosphine,
red phosphorus, and phosphoric acid :

P406 + 6H20 - PH3 + 3H3P04.

Alkalies act similarly. Phosphorus trioxide (or phosphorous an-
hydride) ignites in contact with absolute alcohol : ether, carbon
disulphide, benzene, and chloroform dissolve it without decom-
position. With ammonia, it forms the diamide of phosphorous
acid, HO-P(NH2)2.

Phosphorus tetroxide, P204.— When liquid P4O6 is heated in a
sealed tube it is stable up to 200° ; at 210° it becomes turbid, and
at 290° a sublimate of phosphorus tetroxide, P204, and a residue of
red phosphorus are formed : 2P(1O6 = 3P204 -f 2P. The tetroxide
sublimes in vacuo at 180°. If phosphorus is burnt in a tube in a

XXXT

PHOSPHORUS (539

limited supply of air, a buff-coloured powder is deposited on the
cooler part, which consists of a mixture of P4010, P406, and red
phosphorus. On heating this in a sealed tube, a white crystalline
sublimate of P204 is formed : P2O3 -f P2O5 *= 2P204. With water,
this gives a mixture of phosphorous and phosphoric acids :
P204 -f 3H20 = H3P03 + H3P04.

An oxide, P2O, is said to be formed as a reddish-yellow powder by the
action of the silent discharge on a mixture of hydrogen and POC13
vapour, by the action of PH3 on POBr3, or by heating phosphorous acid
and POC13 : 2H3PO3 = 3H2O + P2O3 ; 2P2O3 - P2O5 + P2O.

Phosphorous acid, H3P03. — Phosphorous acid, H3P03, is formed
when the trioxide is dissolved in cold water, but is most conveniently
prepared by the action of water on its acid chloride, phosphorus
trichloride (Davy, 1812) : PC13 + 3H2O = H3P03 -f 3HC1.

To minimise the decomposing action of the rise of temperature
produced, a stream of air may be passed through the trichloride and the
vapour passed into ice-cold water. Or the trichloride may be added to
concentrated hydrochloric acid, when gaseous hydrogen chloride is
evolved, and the heat of reaction is then diminished by the heat ab-
sorbed in the evolution of hydrochloric acid gas from the solution. The
formation and decomposition of the trichloride may be carried on simul-
taneously by passing chlorine through phosphorus melted under water.

The solution is evaporated until the temperature rises to 180°,
hydrogen chloride being driven off. and it then crystallises on
cooling. The crystalline acid is also obtained by heating PC13
with oxalic acid until frothing ceases, and then cooling :

PC13 + 3C2H204 = H3P03 4- 3C02 + 3CO -f 3HC1.
Phosphorous acid forms white crystals, m.-pt. 71-7°; it is very
soluble in water. When heated it decomposes, evolving pure
phosphine, and leaving metaphosphoric acid : 4H3P03 =
3HP03 + 3H20 -f PH3. If the acid is heated in the air, the
phosphine ignites and burns in bright flashes. This result is
obtained by heating the residue obtained by burning phosphorus
in a confined volume of air over water ; this contains phosphorous
acid, formed from phosphorus trioxide. Phosphorous acid is a powerful
reducing agent, precipitating many metals, such as gold, from solutions
of their salts, and it reduces mercuric to mercurous chloride :

2HgCl2 + H20 + H3P03 - Hg2012 (pp.) + 2HC1 + H3P04.
Silver nitrate gives first a white precipitate of phosphite, Ag3PO3,
which rapidly turns black from formation of metallic silver. Phos-
phorous acid precipitates sulphur from a solution of sulphurous acid :
H2SO3 -f 2H3PO3 = 2H3P04 + H2O + S ; it is slowly oxidised by
solutions of iodine and potassium permanganate.

640 IXOIUJAN'U' CHEMIST UY < HAI>.

Wurtz found that phosphorous acid, although it has the formula
H3PO3, is dibasic; only two atoms of hydrogen can bo replaced by
metals to form salts. Its preparation from phosphorus trichloride
points to the formula P(OH);] :

/;(TH!-OH /OH

/ ms

P— ;C1 H;-OH -> P-rOH+3HCl.

N(l Hi -OH \)H

To explain its dibasic character, however, the formula of the

V/OH

acid is usually written : O=P — OH, the hydroxyl hydro^eii atoms

being ionised in solution, whilst the hydrogen atom directly attached
to the phosphorus (which is quinquevalent) is not split off as an
ion (cf. p. 517). It is supposed that, although the formula may be
P(OH)3 at the instant of its formation from the trichloride, the
molecule of the acid undergoes almost immediate internal rearrange-
ment, or tautomeric change (p. 497) :

/OH /OH

in/ v/

P— OH =z± 0=P— OH.

\OH \H

The existence of two isomeric ethyl phosphorous acids, which max be
/H XC2H5

written O:P — OH and O:P — OH , supports this hypothesis. Normal

\OC2H5 \OH

esters of phosphorous acid, e.g., ethyl phosphite, P(OC2H5)3, are, however,
also knowrn, and the dibasic character of the acid may simply be due to
the increasing difficulty of splitting off hydrogen ions in the successive
dissociations : H3PO3 =r H' +H2PO3'^± H' +H' +HPO3" -~ 3H' +PO3"'.
The reducing properties of the acid, however, appear to be due to the
hydrogen atom directly attached to phosphorus.

The two series of salts known are RH2P03 and R2HP03. When
boiled with alkalies, they evolve hvdrogen : H3PO3 -f- 3KOH =
K3P04 + 2H20 + H2.

The acid H3PO3 is orthophosphorous acid. Pyrophosphorous acid,
H4P2O5, or, since it is dibasic, H2(H2P2O5), is formed by shaking PC13
with H3PO3 for five hours at 30-40°, and leaving in a desiccator over
P2O5 ; it forms needles, m.-pt. 38°. Met a phosphorous acid, HPO2, is
formed in crystals by the oxidation of phosphine by oxygen under
25 mm. pressure : PH3 -f- O2 = HPO2 + H2. When phosphine is
exploded with oxygen, this reaction occurs, together with the reaction
2PH3 + 302 = 2H3PO3.

\\\l

PHOSPHORUS

641

Phosphorous acid reacts with phosphorus pentachloride in the
normal manner, forming the acid chloride, PC13:H,PO« + 3PCL •

PCI, i .*{ pod., | :$nri.

Hypophosphoric acid, H2PO:,.— If sticks of phosphorus, enclosed
in i^lass tubes open at both ends (Ki^. 323), are supported in a glass
funnel over a beaker of water under a bell-jar, oxidation occurs,
with the production of fumes, which sink into the beaker and dis-
solve, rendering the water acid. Dulong first noticed that the
acid made in this way, called " Pelletier's phosphorous acid "
(I7!)f>), differed from ordinary phosphorous acid; he called it
phosphatic acid. Salzcr (1877)
found that if the liquid is
partially neutralised with
caustic soda, sparingly soluble
crystals of the composition
NalIPO3,3H2O slowly separate
from the acid liquid. If lead
nitrate is added to a solution
of this salt, the lead compound,
PbP03, is precipitated ; on
suspending this in water and
passing sulphuretted hydrogen,
a solution of the free acid, now

usually called hypophosphorio FlQ. 323.— Preparation of Hypophosphoric Acid.

acid, H2P03, is obtainea. This

on evaporation in a vacuum desiccator over sulphuric acid gives

crystals, H2PO8,H20, which readily lose water and give H2P03.

If phosphorus is heated on a water-bath with a solution of copper
nitrate, or an acid solution of silver nitrate, copper or silver phosphides,
and thon salts of hypophosphoric acid, are formed. Six gm. of silver
may be dissolved in 100 gm. of nitric acid diluted with its own volume
of water, and 9 gm. of white phosphorus added. When the violent
reaction which occurs on heating subsides, the solution is cooled, and
silver hypophosphate, Ag2PO3, separates ; this may be decomposed by
Hydrochloric acid to obtain hypophosphoric acid.

Hypophosphorio acid on heating decomposes with evolution of
phosphine, leaving phosphoric acid. It differs from phosphorous
acid in having no reducing action on metallic salts. The hypo-
phosphates are oxidised by bromine water to pyrophosphates,
which indicates that the formula of the acid is :

/OH
O = P<
I N)H
yOH

0 = P

T T

642 INORGANIC CHEMISTRY CHAP.

The simpler formula, H2PO3, is now adopted, since the vapour
density of the ester shows that it has the formula (C2H5)2P03. The
existence of an acid salt, Na3HP206,9H2O, may be cited in
evidence of the formula H4P2O6, although this salt may be
3Na2PO3 -f- H2P03 + 18H2O. Hypophosphoric acid is rapidly
oxidised by potassium permanganate, but phosphorus tetroxide,
P2O4, is only slowly oxidised. This oxide, with water, produces
only a mixture of phosphoric and phosphorous acids : it is probably
not the true anhydride of hypophosphoric acid, but is phosphoryl

phosphate, O:P-0-Pf .
^0

Hypophosphorous acid, H3P02. — This acid was discovered by
Dulong in 1816. The residue from the preparation of phosphine
from phosphorus and alkali (p. 618) contains a salt of hypophos-
phorous acid, H3P02. The acid is prepared by warming white
phosphorus with baryta water :

2P4 + 3Ba(OH)2 + 6H20 = 2PH3 + 3Ba(H2PO2)2.

The solution is filtered from barium phosphate also formed, the
excess of baryta is removed by carbon dioxide, and the barium
hypophosphite, Ba(H2P02)2,H2O, recrystallised. A solution of
barium hypophosphite is then decomposed with the calculated
amount of sulphuric acid : •

Ba(H2P02)2 -f H2S04 - BaS04 + 2H3P02.

The filtrate is carefully evaporated, below 130°, to a syrup, cooled
to 0° in a desiccator over P205 and KOH, and crystallised. The
acid is also formed by passing carbon dioxide saturated with the
vapour of phosphorus trichloride into a paste of water and phos-
phorus trioxide.

Hypophosphorous acid melts at 17 '4° ; on heating it decom-
poses at 130°, becoming yellow, and evolving phosphine : 4H3PO2 =
2HP03 + 2PH3 -f 2H26. The salts also evolve phosphine on
heating :

4NaH2P02 = 2PH3 + 2Na2HPO4 = 2PH3 -f Na4P207 -f H20.

Hypophosphorous acid and its salts are powerful reducing agents,
precipitating metals from solutions of their salts. Thus, silver
nitrate gives a black precipitate of silver. From copper salts,
cuprous hydride, CuH, is thrown down, which evolves hydrogen on
warming with hydrochloric acid. The acid is monobasic, forming
crystalline salts such as sodium hypophosphite, NaH2P02,H2O, and
calcium hypophosphite, Ca(H2PO2)2. These are prepared by boiling
phosphorus with caustic soda, or milk of lime, respectively, and are

xxxi PHOSPHORUS 643

used medicinally as tonics. Since it is monobasic, the acid is

usually given the formula O:P — OH .

The hydrogen atoms directly attached to phosphorus have re-
ducing properties (cf. phosphorous acid). All hypophosphites are
soluble in water. The acid is reduced by zinc and hydrochloric
acid to phosphine.

EXERCISES ON CHAPTER XXXI

1. Give a general account of the properties of the elements of the
nitrogen group, with special reference to the change of properties with
increase of atomic weight.

2. What are the chief minerals containing phosphorus ? How is the
element prepared on the large scale ? How is it purified, and for what
purposes is it used ?

3. Discuss the valency of phosphorus in its compounds. It was once

assumed that PC15 was a " molecular compound," PC13,C12. What
facts make this improbable ?

4. What happens when (a) phosphorus is boiled with caustic soda
solution, (b) phosphorous acid is heated, (c) chlorine is passed through
phosphorus fused under water, (d) silver nitrate is added to a solution
of ordinary sodium phosphate ? Give equations.

5. Describe the allotropic modifications of phosphorus. How
may red phosphorus be obtained from yellow phosphorus, and vice
versa ?

6. Describe briefly the preparation and properties of the hydrogen
compounds of phosphorus. Compare their properties with those of
nitrogen.

7. What is the action of water on (a) calcium phosphide, (b) phos-
phorus trichloride, (c) phosphorus tri-iodide, (d) phosphorus penta-
chloride, (e) metaphosphoric acid ?

8. Tabulate the properties and reactions of: (1) yellow and red
phosphorus ; (2)ortho-, pyro-, and meta-phosphoric acids, so as to show
their differences.

9. You are given a solution which may contain a phosphite, a hypo-
phosphite, or a hypophosphate. Explain carefully, giving equations,
how you would distinguish between salts of these acids.

10. How are the lower oxides of phosphorus prepared ? What is the
action of water upon them ?

11. Describe the preparation and properties of the sulphides of
phosphorus, and thiophosphoryl chloride.

T T 2

CHAPTER XXXII

ARSENIC AND ITS COMPOUNDS

Arsenic. — The two minerals realgar (red), As2S2, and orpiment
(yellow), As2S3, were known to the ancients, but were confused
with cinnabar (HgS), under the name sandarach, or arsenicon. The
Greek alchemist Olympiodorus (fifth century) describes the pro-
duction of white arsenic (arsenious oxide, As406) by roasting the
sulphides in. air ; he calls it " white alum." The element arsenic
itself, obtained as a sublimate, was also known, and used for whiten-
ing copper, forming an alloy with the metal, and was thence regarded
as a " second mercury." Arsenical compounds, which are very
poisonous, were introduced into medicine by Paracelsus in the
sixteenth century (cf. p. 29). The composition of white arsenic,
as the calx of ' metallic " arsenic, was recognised by Brandt in
1773.

The chief minerals containing arsenic are the sulphides,
orpiment (As2S3) and realgar (As2S2) ; the oxide, arsenite,
As4O6 ; arsenical iron, FeAs2 ; arsenical nickel, NiAs ; nickel
glance, NiAsS ; tin-white cobalt, (Co,Ni.Fe)As2 ; arsenical pyrites,
or mispickel, FeAsS ; cobaltite, CoAsS ; and certain oxidised
compounds containing salts of arsenic acid, H3As04 ; pharmacolite,
(CaHAs04)2 + 5H20 ; cobalt bloom, Co3(As04)2 -f 8H20 ; and
mimetisite, 2Pb3(As04)2,Pb2(P04)Cl. The free element is also found
in large quantities.

Iron pyrites and other sulphide ores often contain arsenic, which
appears to replace sulphur, and function as a bivalent element :
Fe(As,S)2. Sulphuric acid prepared from arsenical pyrites may
contain 1 per cent, of As203 (p. 508), and coal smoke, especially in
yellow fogs, may contain arsenious oxide, from the pyrites in the
coal. Traces of arsenic occur in nearly all materials, including the
human body, and most foods. About 10,000 tons of arsenic com-
pounds are produced annually, mostly in the Freiburg Smelteries.

Arsenious oxide, As203. — In roasting minerals in a current of air,
for metallurgical treatment, fumes of arsenious oxide, As406 (or,
since the molecular weight of the solid is unknown, As2O3), are often
evolved, and may be condensed in flues as an impure powder :
e.g., 4CoAsS + 9O2 = 4CoO + 4SO2 + 2As2O3. This may be ob-

644

CH. XXXII

ARSENIC AND ITS COMPOUNDS

645

tained in larger quantities by roasting rich arsenical ores, such as
mispickel.

A suitable apparatus is the revolving calciner of Oxl and and
Hocking (Fig. 324), consisting of a rotating iron cylinder lined with
refractory material, down which the crushed ore slides from a
hopper above. The ore is met by flames and hot gases from a
furnace at the lower end. The " arsenical soot " is collected in
flues, and the roasted ore, freed from arsenic, drops into a wagon
for use in the smelting furnace.

The impure arsenious oxide may be purified by sublimation in
iron pots, and the while arsenic so obtained is the source of all the
arsenic compounds of commerce. The production at present
exceeds the demand.

Arsenious oxide is used as a poison for vermin, in taxidermy for
preserving skins, and in glass-making for removing colour from

/ m

FIG. 324. — Oxland and Hocking's Revolving Calciner.

the glass. In minute amounts it is used medicinally as a tonic,
and in diseases of the skin.

Arsenic. — The element (" metallic -arsenic ") occurs "native, and
is also obtained by heating arsenical pyrites with iron, or by reducing
the trioxide with charcoal. The powdered mixture is heated in a
clay crucible, covered with an inverted iron cone, into which the
arsenic sublimes : As406 + 6C = 6CO -+- As4. It is also prepared
on the larger scale by heating mispickel in a clay tube, fitted for
half its length with an inner tube of sheet iron. The iron tube is
afterwards unrolled to split off the arsenic : FeAsS = FeS + As.
Arsenic is purified by resublimation from charcoal powder. Arsenic
sulphides are not reduced by heating with charcoal ; with potassium
cyanide, they give arsenic.

EXPT. 261. — Heat a little arsenious oxide with powdered charcoal
and potassium cyanide in a dry test-tube. A black mirror of arsenic
sublimes in the tube. If this is heated, it is oxidised, and a white
sublimate of arsenious oxide forms higher up in the tube.

Allotropic forms of Arsenic. — As in the case of phosphorus (p. 614),
the element exists in different allotropic forms.

646 INORGANIC CHEMISTRY CHAP.

The following modifications of arsenic have been described :

(1) a- Arsenic, or yellow arsenic, corresponding with yellow phos-
phorus ; soluble in carbon disulphide ; an unstable form ; sp.
gr. 3-7 (Schuller, 1889).

(2) ft- Arsenic, or black arsenic, sp. gr. 4-7; less stable than y-arsenic;
insoluble in carbon disulphide (Retgers, 1893).

(3) y- Arsenic, or grey arsenic ("metallic arsenic"), the stable and
ordinary form, corresponding with " metallic " phosphorus ;
sp. gr. 5-73 ; insoluble in carbon disulphide.

Grey arsenic forms steel-grey, brittle, hexagonal-rhombohedral
crystals, with a metallic lustre, which are fairly good conductors
of heat and electricity. It volatilises slowly at 100° ; at 450° it
sublimes rapidly, without previous fusion, forming a lemon-yellow
vapour, the density of which varies with the temperature, indicating
the dissociation : As4 ;=± 2As2.

Temp. 860° 1714° 1736°

A (H = 1) 147 79 77

(Theoretical : As4 150 ; As2 75.)

In respect of its molecular weight, arsenic resembles phosphorus,
and differs from the metals, which are usually monatomic. When
heated in a sealed tube under pressure, grey arsenic melts at 480°.

Grey arsenic is not oxidised in dry air, but in presence of moisture
it rapidly becomes covered with a blackish-grey film, containing
the trioxide. When heated to 200° in air it shows a distinct
phosphorescence ; at 400 ° it burns in air with a white flame. The
element burns brilliantly in oxygen : As4 -f- 302 = As4O6.

EXPT- 262,. — Heat 1 gm. of arsenic in a current of oxygen in a hard
glass tube connected with an empty flask, the exit tube from which
passes to a U -tube packed with glass wool to keep back arsenious oxide.
The arsenic burns with a brilliant flame, producing white solid arsenious
oxide.

Powdered arsenic takes fire in chlorine, forming the trichloride.
Arsenic combines with most metals to form fusible arsenides ;
0-3-1 per cent, of arsenic alloyed with lead makes the latter
harder and more fusible. If this alloy is allowed to flow through
a sieve, the drops of fused metal, falling down a tower into water,
assume a spherical shape, and form shot.

Hydrochloric acid dissolves arsenic only in the presence of air :
the trioxide is probably first formed :

As406 + 12HC1 = 4AsCl3 + 6H20.

Dilute nitric acid has little action in the cold ; the hot dilute acid
slowly oxidises arsenic to arsenic acid, H3As04, and this is rapidly
formed with concentrated nitric acid, or aqua regia. Hot concen-

xxxn ARSENIC AND ITS COMPOUNDS 647

trated sulphuric acid is reduced to sulphur dioxide, and an
unstable arsenious sulphate, As2(S04)3, appears to be formed, but
decomposes into the oxide. Arsenic is insoluble in alkalies.

/3- Arsenic, or black arsenic, is formed when ordinary arsenic is rapidly
heated in a glass tube in a current of hydrogen, when the element
volatilises, and is deposited on the cold tube further on, partly in grey
rhombohedral crystals of y-arsenic, near the heated portion of the tube,
and partly as a black, shining, amorphous deposit of /3-arsenic in the cooler
portion (210-220°). (In the cold part of the tube a grey crystalline
deposit often appears, which may be a fourth form.) /3-arsenic is not
appreciably oxidised by air even at 80°. On heating to 360° it forms
y-arsenic, and may be simply a physical modification of the latter.

Yellow arsenic, or a-arsenic, is a peculiar allotropic form, resem-
bling yellow phosphorus, produced by rapidly cooling arsenic
vapour. Arsenic is distilled in a current of carbon dioxide, and
the gases are passed into a U-tube, where they meet a current of
cooled carbon dioxide. The gases are then led into cold carbon
disulphide, which dissolves the a-arsenic (8 gm. in 100 c.c. at 20°).
On evaporation, light yellow regular crystals are deposited, which
rapidly oxidise in the air at the ordinary temperature with a fault
luminescence and a garlic odour, thus behaving like yellow phos-
phorus. On exposure to light, even at — 180°, they rapidly pass
into y-arsenic. Yellow arsenic is also formed quantitatively by
volatilising y-arsenic in vacuo and cooling with liquid air. Its
molecular weight in solution in carbon disulphide corresponds

As=As

with As 4 ; the formula may be |

As=As

By the action of stannous chloride on a solution of arsenious
oxide a brown precipitate of arsenic is formed, part of which is
soluble in carbon disulphide, and consists of a-arsenic. The pro-
portion of the latter is increased if the mixture is shaken with
carbon disulphide during the reduction, since the solution of
a-arsenic is more stable than the solid.

Arsenic trihydride, AsH3. — The only hydrogen compound of
arsenic definitely known is the trihydride, AsH3, called arsine, or
arseniuretted hydrogen. It is a colourless gas, b.-pt. — 55°, m.-pt.
- 119°. It is not formed by direct combination of the elements,
but is produced by the action of nascent hydrogen on a soluble
arsenic compound. Thus, if a solution of arsenious oxide is added
to a mixture of zinc and sulphuric acid which is evolving hydrogen,
or to sodium amalgam, the gas acquires a very unpleasant smell of
garlic, is extremely poisonous, and burns with a green flame (Scheele,
1775). It is also formed at the cathode by the electrolysis of solu-

648 INORGANIC CHEMISTRY CHAP.

tions of arsenious oxide and by boiling a soluble arsenic compound
with zinc and caustic potash ; antimony does not form a hydride
in the latter reaction (Fleitmann, 1850). The gas obtained by all
these processes is largely diluted with hydrogen. If it is passed
through a tube cooled in liquid air, the arsine is liquefied, and on
warming the pure gas is evolved.

Pure arsine may also be prepared by the action of dilute sulphuric
acid on zinc arsenide, Zn3As2, prepared by heating arsenic and zinc
in a crucible : Zn3As2 + 6HC1 = 2AsH3 -+- 3ZnCl2 ; by the action
of water on sodium arsenide, which is formed by passing the impure
gas over heated sodium : Na3As + 3H20 = AsH3 -f SNaOH ; by
heating sodium formate (dried at 210°) with sodium arsenite ; or,
most conveniently, by the action of warm water on aluminium
arsenide, obtained by heating together aluminium powder and
powdered arsenic in a covered crucible : AlAs -j- 3H2O =
A1(OH)3 -f- AsH3. The gas is exceedingly poisonous, although
from the experience of the author this property seems to have been
somewhat overrated. By the growth of moulds hi presence of
arsenic compounds (e.g., Scheele's green in wall-paper), ethyl arsine,
AsH2C2H5, is formed ; this smells of arsine, and is poisonous.

On exposure to light in the moist condition, arsine is rapidly
decomposed, with deposition of a black, shining deposit of arsenic
on the side of the jar ; a little yellow arsenic is usually formed.
The gas is decomposed by heat into its elements, the reaction com-
mencing at about 230° : after decomposition, the volume of the
gas increases in the ratio 3 : 2 — 2AsH3 = 2As + 3H2. From this
result, and the density, the formula of the gas is found.

Arsine differs from ammonia and resembles phosphine in being
almost insoluble in water. Unlike phosphine, it is almost insoluble
in alcohol ; it is nearly insoluble in ether, but dissolves readily in
turpentine.

The Marsh-Berzelius test. — The formation of a gaseous hydride
and its ready decomposition by heat form the basis of the very
delicate Marsh-Berzelius test. Since, if a soluble arsenic compound
is added to zinc and acid evolving hydrogen, the whole of the arsenic
is ultimately expelled as hydride, the test may be used quanti-
tatively.

EXPT. 263. — Hydrogen is generated in a flask from, pure (electrolytic)
zinc and pure dilute sulphuric acid ; the gas is freed from traces of
sulphuretted hydrogen by a roll of dry lead acetate paper in the first
part of the drying tube, the second half of which is packed with pure
granular calcium chloride, separated from the paper by a plug of cotton-
wool (Fig. 325). The dry gas then passes through a hard glass tube,
constricted as shown, and heated at one point to dull redness by a
Bunsen flame. If the materials are free from arsenic, no stain is pro-

xxxii ARSENIC AND ITS COMPOUNDS 649

dueed in this tube beyond the heated portion. If now a few drops of
a dilute solution of arsenious oxide, or any material to be tested for
arsenic, are added to the flask, arsine is formed, which is decomposed in
the hot tube, a brown or black mirror being deposited beyond the
heated portion. After a sufficient time, the whole of the arsenic is
expelled from the solution as arsine, and by comparing the mirror with
standard tubes prepared
with known amounts
of arsenious oxide
(0-001-0-01 mgm.) a
quantitative estimation
may be made.

If the tube is not
heated, but the gas
kindled at the jet, the
flame, which is tinged
green, deposits black
spots of arsenic on the
outer surface of a glazed
porcelain dish filled with
water. These are pro-
duced by decomposition
by the heat of the
flame : 2AsH3 = 2As +

3H2 ; they dissolve readily in a solution of sodium hypochlorite or
bleaching powder (forming arsenates, e.g., Na3AsO4), but are insoluble
in tartaric acid. If a spot is moistened with yellow ammonium sul-
phide, and this evaporated by gentle heating, a bright yellow spot of
arsenic trisulphide, As2S3, is left (c/. antimony hydride, p. 938).

If arseniuretted hydrogen is passed into dilute silver nitrate
solution, a black precipitate of metallic silver is formed, and the
filtrate contains arsenious acid (cf. antimony, p. 939) :

2AsH3 + 12AgN03 + 6H2O = 2H3As03 + 12HNO3 + 12Ag.

If the solution of silver nitrate is more concentrated, no pre-
cipitate is formed, but a yellow solution of a double compound
of silver arsenide and nitrate is obtained :

AsH3 + 6AgN03 = Ag3As-3AgN03 + 3HN03.

On dilution with water, a black precipitate of metallic silver is
deposited :

Ag3As-3AgNO3 + 3H2O = 6Ag + 3HN03 + H3AsO3.

If the gas is passed into mercuric chloride solution, a yellow
coloration is produced, due to the formation of AsH(HgCl)2 ; on

FIG. 325.— Marsh-Berzelius Test for Arsenic.

650 INORGANIC CHEMISTRY CHAP.

further treatment, this gives brown As(HgCl)3, and finally black
As2Hg3. This is the basis of the Gutzeit test.

The liquid is added to zinc and dilute sulphuric acid in a test-tube. A
rell of lead acetate paper is placed in the tube to absorb H2S, and a piece
of filter -paper, soaked in mercuric chloride solution and dried, is stretched
over the open mouth of the tube by a rubber band. The yellow stain is
compared with standard stains produced with known amounts of arsenic.
The test is very sensitive.

An ill-defined brown solid hydride is said to be formed by the action of
water on sodium arsenide, by the action of the silent discharge on arsine,
or by the electrolysis of dilute sulphuric acid with a cathode of arsenic.
At 200° it is converted into grey arsenic. It has been given the
formulae As4H2, AsH2, and (AsH)x, but may be impure yellow arsenic.
No definite hydride corresponding with N2H4 or P2H4 is known, but the
organic compound cacodyl, As2(CH3)4, is of this type.

Halogen compounds of arsenic.— The stable halogen compounds of
arsenic, including the fluoride, are of the type AsR3 (cf. phosphorus).

Arsenic trifluoride, AsF3, a colourless fuming liquid, b.-pt. 60-4°,
m.-pt. — 8*5°, sp. gr. 2-66, is prepared by heating a mixture of arsenious
oxide, powdered fluorspar, and concentrated sulphuric acid in a lead
retort : As2O3 -f 6HF = 2AsF3 + 3H2O. The water produced in the
reaction is retained by the sulphuric acid, otherwise hydrolysis of the
fluoride would occur : 2AsF3 + 3H2O ^± As2O3 + 6HF. Arsenic
pentafluoride, AsF5, is obtained as a colourless gas, b.-pt. — 53°, m.-pt.

— 80°, by distilling a mixture of the trifluoride, antimony penta-
fluoride, and bromine at a temperature not exceeding 55°, and collecting
in a receiver cooled in liquid air : AsF3 + 2SbF5 + Br2 = AsF5 -f-
2SbBrF4. The double salts K2AsFr,H2O and KAsOF4,H2O are
formed as crystalline solids when potassium arsenate, K3AsO4, is
dissolved in hydrofluoric acid.

Arsenic trichloride, AsCl3, discovered by Glauber (1648), is the
most important halogen compound of arsenic. It is formed when
arsenic burns in chlorine gas — a reaction which occurs spon-
taneously even if the materials are very carefully dried — but is usually
prepared by distilling a mixture of white arsenic, common salt,
and concentrated sulphuric acid in a retort, and condensing the
vapour in a cooled receiver : As203 + 6HC1 = 2AsCl3 + 3H2O.
The water is removed by the excess of sulphuric acid, otherwise
hydrolysis would occur. The distillate is freed from excess of
chlorine by distillation over powdered arsenic.

Arsenic trichloride is a colourless, oily liquid, b.-pt. 130°, m.-pt.

— 18°, sp. gr. 2-2, which fumes in moist air, and is rapidly hydro-
lysed by water : the first product is a crystalline hydroxychloride,

xxxn ARSENIC AND ITS COMPOUNDS 651

but with excess of water, arsenious acid (or its anhydride, As406)
is formed :

(1) AsCls 4- 2H20 = 2HC1 -f AsCl(OH)2 ;

(2) AsCl(OH)2 + H20 = As(OH)3 + HC1.

The hydrolysis is reversible to a slight extent, showing that arsenic
is approaching a metal in its properties (p. 450) ; if arsenious oxide
is dissolved in hydrochloric acid and the liquid boiled, arsenious
chloride distils over with the steam : As4O6 -j- 12HC1 ^
4AsCl3 + 6H20.

Arsenic oxychloride, AsOCl, is formed as a colourless fuming liquid by
adding arsenic trioxide to the boiling trichloride. When heated, it
gives AsCl3 and a compound As3O4Cl. With water, it forms AsCl(OH)2.

Arsenic pentachloride, AsCl6, is said to be formed from the trichloride
and chlorine at — 40°, but decomposes into its constituents above
— 25°, and may be simply a solution of chlorine in the trichloride.

Arsenic tribromide, AsBr3, is a colourless crystalline solid, m.-pt.
31°, b.-pt. 221°, less easily hydrolysed than A*C13, and arsenic tri-iodide
forms red hexagonal crystals ; both compounds are formed by heating
arsenic with a solution of the halogen in carbon disulphide. The tri-
iodide is only slightly hydrolysed by water, and is formed on adding- a
solution of arsenious oxide in hot hydrochloric acid to a solution of
potassium iodide. A di-iodide, AsI2, is obtained as a dark red mass
by heating iodine with arsenic in a closed tube to 260° ; it is soluble in
carbon disulphide, but is decomposed by water into AsI3 and arsenic.
By heating AsI3 with iodine to 150° a brown pentaiodide, AsI5, is said
to be formed. A brown mono-iodide, Asl, is produced as a brown
powder when an alcoholic solution of iodine is saturated with arsenic.

Arsenious oxide. — Arsenious oxide, or arsenic trioxide, As406
(usually written As203), is the most important compound of arsenic ;
it is known in commerce as " white arsenic," or simply as " arsenic."
It was known to the ancients, and used as a caustic. The intensely
poisonous properties of the substance were first recognised by
Paracelsus, doubtless from the results of his reckless use of arsenic
as a medicine, and it was a favourite poison during the Middle
Ages — the aqua tofani. It exists in three varieties : — (1) the
amorphous, sp. gr. 3-738, m.-pt. 200° ; (2) the octahedral, sp. gr.
3-689, sublimes without fusion ; (3) the rhombic, sp. gr. 3-85, occur-
ring as the mineral daudetite.

The vapour density of arsenious oxide between 570° and 1560°
corresponds with the formula As406 ; at 1770° it is As203. In
solution in nitrobenzene the formula is also As4O6.

The amorphous variety is formed as a colourless transparent
glass, first described by Roger Bacon, when the vapour is slowly
condensed at a temperature slightly below its point of vaporisation,
218°. It may be preserved in sealed tubes, but at 100°, or in pre-

652 INORGANIC CHEMISTRY CHAP.

sence of moisture, it becomes opaque, and passes into the octahedral
form. If the glass is dissolved in concentrated hydrochloric acid
and the solution allowed to cool, crystals are 'deposited, each accom-
panied by a flash of light. The octahedral form is said not to
exhibit this property (H. Rose). The vitreous form dissolves in
about 25 parts of water at 10°, or in 12 at 100°, but the solubility
diminishes on standing, owing to conversion into the octahedral
form. The latter dissolves in about 70 parts of water at the
ordinary temperature, but exceedingly slowly.

The octahedral is the stable form under ordinary conditions ; it
is produced when the vapour is rapidly condensed, when the tri-
oxide is crystallised from water or hydrochloric acid, or spon-
taneously, with evolution of heat, from the vitreous form. It
sublimes at 125-150°, but can be fused under increased pressure.

The rhombic variety is formed by crystallisation from a boiling
saturated solution of the amorphous substance in caustic potash,
or when the other varieties are heated for some time at 200°.

If arsenious oxide is heated in a sealed tube to 400°, the vitreous
form remains at the bottom of the tube, the rhombic form sublimes to
the intermediate part, and the octahedral form sublimes to the top of
the tube. The different crystalline forms may be recognised under the
microscope.

Arsenious oxide is easily oxidised to arsenic oxide, or arsenic
pentoxide, As2O5, by ozone, hydrogen peroxide, chlorine, aqua
regia, bromine,' iodine, nitric acid, and hypochlorites (especially
in alkaline solution) ; when arsenic acid or an arsenate is formed :
As203 + 2C12 + 2H2O = As205 -f 4HC1. It precipitates red cuprous
oxide from Fehling's solution (p. 815).

Arsenious oxide is also easily reduced to arsenic by heating in a
tube with charcoal, when a mirror of arsenic sublimes, or by a
solution of stannous chloride, which gives a brown precipitate :
As2O3 + 3SnCl2 -f- 6HC1 = 3SnCl4 + 2As + 3H2O. If arsenious
oxide is boiled with hydrochloric acid and copper foil, the latter
becomes grey, owing to deposition of arsenic :

As2O3 + 6HC1 + 6Cu = As + 6CuCl + 3H2O.

If the copper foil is now washed, dried, and heated in a tube, a
crystalline sublimate of arsenious oxide is formed (Reinsch's test).

By the action of fuming sulphuric acid on the trioxide, unstable sul-
phates, composed of As2O3 with 1, 2, 3, 4, 6, and 8 SO3, are formed :
As2O3 then acts as a feebly basic oxide. These are decomposed by
water.

Small quantities of arsenious acid occur in some mineral waters,
which are used as nerve tonics, and in improving the blood. Arsenious
oxide is a violent poison : 0-06 gm. is a dangerous dose, and 0-125-0-25

xxxn ARSENIC AND ITS COMPOUNDS 653

gm. is fatal. Habitual use of small quantities renders the system
immune to much larger doses, and the peasants of Styria are able to
consume arsenious oxide in amounts (0-3 gm.) which would be fatal to
those unaccustomed to its use. It is said by them to act as a cosmetic,
to improve the breathing in mountain climbing, and to give plumpness
to the figure. Freshly precipitated ferric hydroxide, obtained by adding
magnesia to a solution of ferrous sulphate, absorbs arsenious oxide, and
is the best antidote in cases of poisoning

Arsenious acid. — A solution of arsenious oxide in water has a
feebly acid reaction ; it probably contains arsenious acid, H3As03,
or HAsO2, although only the trioxide crystallises on cooling or con-
centration : As203 + 3H2O ±=; 2As(OH)3. The acid is even weaker
than sulphuretted hydrogen.

The finely-powdered oxide is not easily wetted by water, but a solution
can be prepared by boiling. It also dissolves in warm sodium bicar-
bonate solution, with evolution of carbon dioxide, and formation of
sodium arsenite, Na3AsO3, or NaAsO2. This solution is often used for
the standardisation of iodine solution. The latter oxidises the arsenite
to arsenate : As2O3 + 2I2 + 2H2O = As^ + 4HI. The excess of
bicarbonate has no action on the iodine, whilst if the arsenious oxide
were dissolved in caustic alkali, the latter would react with iodine.

Three series of arsenites appear to exist, derived from the hypo-
thetical acids :

O^oarsenious acid, H3As03 ; e.g.,

K3AsO3, Ag3As03 ; HCuAs03, Pb3(As03)2.
P?/roarsenious acid, H4As205 ; e.g., Ca2As205; (NH4)4As206.

id, HAsO2 ; e.g., KAsO2, Ba(AsO2)2, KH(AsO2)2.

If arsenic trioxide is boiled with caustic alkali, carbonate or
bicarbonate, alkali meta-arsenites, e.g., NaAsO2, are formed.
Arsenic trioxide solution, neutralised with ammonia, gives with
silver nitrate a yellow precipitate of silver arsenite, Ag3AsO3, soluble
in acetic acid (the yellow silver phosphate, Ag3PO4, is insoluble).
Copper sulphate added to the ammonium arsenite solution gives
a bright green precipitate of cupric hydrogen arsenite, CuHAsO3,
known as Scheele's green, and formerly used for colouring wall-
paper. When dissolved in ammonia, this salt is converted into
cuprous ar senate. The brilliant pigment Schweinfurter green, which
has the composition Cu3(As02)2,Cu(C2H3O2)2, i.e~ a compound of
cupric arsenite and cupric acetate, is obtained by boiling verdigris
(a basic acetate of copper) with arsenious oxide and acetic acid.

Arsenic dioxide, AsO2 (or As2O4), is said to be formed as a glass by
heating equimolecular amounts of trioxide and pentoxide to 350°.

654 INORGANIC CHEMISTRY CHAP.

Arsenic pentoxide and arsenic acids. — Unlike phosphorus, arsenic
on combustion in oxygen yields, not the pentoxide, but the trioxide.
The latter may, however, be converted into arsenic pentoxide,
As2O5, by oxidising agents (p. 379). It was prepared by Scheele
(1775), by boiling white arsenic with concentrated nitric acid or
aqua regia : it is formed on heating the residue from the preparation
of nitrous anhydride (p. 587) :

As2O3 -f 2HNOi3 = As2O5 + H2O + N203.

The solution on cooling deposits rhombic crystals of arsenic acid,
2H3As04,H2O. At 100° these melt, lose water of crystallisation,
and leave a crystalline powder of the composition HgAsgG^, or
3As2O5,5H2O. At 160° the acid slowly loses all the contained
water and forms arsenic pentoxide, As2O5, as a deliquescent, white,
crystalline solid. At 200° the water is eliminated in a much
shorter time. The pyro- and meta-acids do not appear to exist,
even in solution, but their salts are known.

Arsenic pentoxide melts at a red heat, and gives off oxygen :
2As205 = 2As203 + 202 (cf. P2O5,N2O5).

The arsenates are isomorphous with the phosphates, and probably
have similar constitutional formulae. The normal orthoarsenates
exist in solution, as well as in the solid state, but the pyro- and
me ta- arsenates exist only in the solid condition, and are prepared
by heating the acid and di-acid ortho-salts, as in the case of
phosphates : 2Na2HAs04 = H00 + Na4As207 ; NaH2As04 =
NaAs03 -f H20. The salt Na2HAs04,12H20 is largely used in
calico-printing. Arsenic acid is an oxidising agent ; e.g., it liberates
iodine from potassium iodide and hydrochloric acid. It was
formerly used in making aniline dyes.

Ammonium molybdate and concentrated nitric acid give with
arsenates a yellow precipitate similar to that obtained with phos-
phates, but only on heating. Magnesia mixture (68 gm. of
MgCl2,6H2O and 165 gm. of NH4C1 dissolved in 300 c.c. of "water,
75 c.c. of ammonia, sp. gr. 0-88, added, and the whole made up to
1 litre) gives a white crystalline precipitate of magnesium ammonium
arsenate, MgNH4AsO4,6H20, similar to MgNH4P04,6H2O. On
heating, this leaves a residue of magnesium pyroarsenate, Mg2As207.

The precipitate of magnesium ammonium arsenate is distinguished
from the phosphate as follows. It is dissolved in dilute hydrochloric
acid, and the hot solution treated with sulphur dioxide. Under these
conditions all arsenates are reduced to arsenites, whilst phosphates are,
of course, unacted upon : As2O5 + 2SO2 = As2O3 + 2SO3. The excess
of sulphur dioxide is removed from the solution by boiling, and a current
of sulphuretted hydrogen passed through the liquid. Yellow arsenious
sulphide is precipitated. The nitrate is boiled to remove H2S, and gives

xxxn ARSENIC AND ITS COMPOUNDS 655

a precipitate of MgNH4PO4,6H2O when made alkaline with ammonia,
if a phosphate is also present. Arsenates are also distinguished from
phosphates by giving with silver nitrate in neutral solution a chocolate-
brown precipitate of silver arsenate, Ag3AsO4, soluble in dilute nitric
acid and in ammonia. Phosphates give a yellow precipitate of Ag3PO4.
If an arsenite is present, it may be detected by dissolving the precipitate
in dilute nitric acid, avoiding excess, and adding ammonia drop by drop .
Brown silver arsenace is first precipitated, then yellow silver arsenite.

Sulphides and thioacids of arsenic. — The trisulphide, As2S3, and
disulphide, As2S2, of arsenic occur native as the yellow and red
minerals orpiment and realgar, respectively. They are prepared
by heating arsenic or arsenic trioxide with sulphur in proper pro-
portions, e.g., 2As203 + 7S = 2As2S2 -f 3S02. The disulphide is
also made at Freiburg by distilling iron pyrites with arsenical
pyrites : 2FeS2 + 2FeAsS = As2S2 + 4FeS. The trisulphide is
easily prepared by passing sulphuretted hydrogen through a solu-
tion of arsenic trioxide in dilute hydrochloric acid :

2AsCl3 + 3H2S = As2S3 + 6HC1.

If sulphuretted hydrogen is passed into a solution of arsenious
oxide in boiling distilled water, no precipitate is formed, but a
yellow colloidal solution of arsenic trisulphide is produced (p. 12).
Addition of dilute hydrochloric acid, or salts, to this at once brings
about coagulation, and yellow flocks of As2S3 separate. If these
are at once filtered off and washed, they again pass into colloidal
solution when the acid or salt has been washed out, but if they
are allowed to stand for some time in the solution in which they
have been precipitated, they become quite insoluble.

Realgar is used in pyrotechny. Bengal fire is a mixture of 27
parts of nitre, 7 parts of sulphur, and 2 parts of realgar. Mixed
with slaked lime, it is used as a depilatory in tanning to remove
hair from hides ; a mixture of orpiment and slaked lime is also
used for removing superfluous hair under the name of " Rusma."
In both cases the active agent is probably calcium hydrosulphide,
Ca(SH)2, which dissolves hair. A mixture of orpiment (the auri
pigmentum of the Romans) with the trioxide, obtained by subliming
the latter with sulphur, is used as a pigment under the name of
King's yellow.

Both sulphides of arsenic burn when heated in air, forming
sulphur dioxide and arsenic trioxide. They are oxidised by nitric
acid, but are insoluble in concentrated hydrochloric acid. (Sb2Ss is
soluble.) Since they are not dissolved by the dilute hydrochloric
acid of gastric juice, the sulphides of arsenic are not poisonous.

Arsenic pentasulphide, As2S5, is said to be formed when sul-
phuretted hydrogen is passed rapidly into a warm solution of
arsenic acid containing 10-12 per cent, of free hydrochloric acid ;

656 INORGANIC CHEMISTRY CHAP.

if the reaction takes place slowly a mixture o£ trisulphide and
sulphur is deposited : As2O5 + 5H2S = As2S3 + 2S + 5H20. The
first product is the unstable thioarsenic acid, H3AsO3S. In quali-
tative analysis, solutions of arsenates are reduced with sulphurous
acid before treating with sulphuretted hydrogen, as the reduction
with the latter is a very slow process.

Arsenic trisulphide dissolves readily in caustic potash, soda, or
ammonia, and even in a warm solution of ammonium carbonate
(antimony trisulphide is insoluble in the latter). The product is
a mixture of an arsenite and a thioarsenite :

As2S3 + 4KOH = K2HAs03 + K2HAsS3 + H2O.

If an acid is added, the whole of the arsenic is precipitated as
sulphide :

K2HAs03 + K2HAsS3 + 4HC1 = 4KC1 + 3H20 + As2S3.

If arsenic trisulphide is dissolved in an alkali-sulphide, a thio-
arsenite alone is formed :

K2S + As2S3 = 2KAsS2 ; or 3K2S + As2S3 = 2K3AsS3.

The compounds (NH4)3AsS3 and Ca3(AsS3)2 form colourless crystals :
K3AsS3 and Na3AsS3 are amorphous white powders. Thioarsenites
are derived from hypothetical thioarsenious acids : H3AsS3 (ortho) ;
H4As2S5 (pyro) ; HAsS2 (meta). Berzelius, who first prepared the
salts, recognised that, in them, sulphur takes the place of oxygen
as the " acidifying principle."

If arsenious sulphide is dissolved in an alkali polysulphide, e.g.,
yellow ammonium sulphide, (NH4)2S2, or if a thioarsenite is digested
with sulphur, a yellow solution of a thioarsenate is obtained :

m v

K3AsS3 + S = K3AsS4. On acidifying the solution, a yellow
precipitate is thrown dcwn, which has been variously described
as the pentasulphide and as a mixture of the trisulphide
and sulphur. Arsenic trisulphide and sulphur, when digested with
caustic potash, form salts containing both oxygen and sulphur ; e.g.,
Na3AsOS3,12H20 ; K3AsOS3 ; Na2HAsOS3,8H2O ; Na3As02S2,HH2O.
These are all colourless, and are decomposed by acids into arsenic
acid and free sulphur, or arsenic trisulphide.

The thioarsenates are soluble, and crystalline ; e.g., Na3AsS4,8H20 ;
(NH4)3AsS4. By the action of sodium sulphide solution on arsenious
oxide in the proportions 2Na2S : As203, a thioarsenate and
elementary arsenic are produced.

EXERCISES ON CHAPTER XXXII

1. What are the chief sources of white arsenic ? How may this be
converted into : (a) arsine, (b) arsenic trichloride, (c) arsenic acid,
(d) arsenic trisulphide ?

xxxn ARSENIC AND ITS COMPOUNDS 657

2. By what reactions is arsenic hydride formed ? How is its forma-
tion used as a test for arsenic ? What other tests for arsenic are applied ?

3. How are arsenites and arsenates prepared ? Compare the proper-
ties of arsenious and arsenic acids with the corresponding acids of
nitrogen and phosphorus. How is arsenic acid distinguished from
phosphoric acid in analysis ?

4. Give examples of oxidising and reducing reactions of arsenious
oxide. What happens if sulphuretted hydrogen is passed through an
aqueous solution of this substance ?

5. In what forms does arsenious oxide exist ? What happens when it
is (a) heated with charcoal, (b) treated in solution with sodium amalgam,
(c) treated with bromine water ?

6. What reactions take place when arsenic trisulphide is (a) boiled
with caustic soda, (b) digested with yellow ammonium sulphide ? If
the products are acidified, what substances are formed ? Give equations.

7. What is the action of water on (a) arsenic trichloride, (6) phos-
phorus pentabromide, (c) arsenic pentoxide ?

8. One hundred c.c. of a gas are collected over mercury in a tube
closed above by a plaster of Paris plug. On standing, diffusion occurs,
and when the mercury level again becomes constant it is found to1 corre-
spond with 164 c.c. What is the molecular weight of the gas ? One
hundred c.c. of the gas on heating with sodium gave 150 c.c. of
hydrogen. What is the gas ?

TT tT

CHAPTER XXXIII

CARBON AND THE HYDROCARBONS

Carbon and its compounds. — The element carbon is found in Nature
in a diversity of forms, both in the free state and in combination.
Elementary carbon occurs in the crystalline forms of diamond, and
graphite (also called plumbago, and black-lead] ; and amorphous as
anthracite coal. Free carbon also occurs in meteorites, and the
spectroscope shows its presence in the cooler stars. In the form
of cyanogen, C2N2, carbon occurs in the sun, and hydrocarbons, or
compounds of carbon and hydrogen, are constituents of some
stars (p. 32). Mixtures of these hydrocarbons compose mineral
oil, or petroleum, which issues from the earth in enormous quan-
tites in Baku (Russia), and Pennsylvania (America). Other oil-
fields are Galicia, Mexico, Burma, Ontario, and Mesopotamia.
Coal contains complex hydrocarbons, but oxygen and nitrogen
are also present. Carbon dioxide, C02, occurs uncombined, e.g.,
in the atmosphere, and as carbonates, especially calcium carbonate,
CaC03 (chalk, limestone, and marble], as magnesium carbonate,
MgCO3 (magnesite], or a compound of the two, CaC03,MgC03,
known as dolomite, of which whole mountain-chains are constituted.

The bodies of plants and animals contain compounds of carbon
with hydrogen and oxygen, and sometimes nitrogen, sulphur, and
phosphorus. • These so-called organic compounds comprise :

(1) Carbohydrates such as starch, C6H10O5 ; various sugars, such as
glucose, or grape-sugar, C6H12O6, and saccharose, or cane-sugar,
C12H22OU ; and cellulose, C6H10O5, or woody-fibre, all occurring in
plants ; (2) proteins, such as albumin, gelatin, and a number of very
complicated compounds occurring both in plants and animals, which
contain nitrogen, and usually sulphur and phosphorus, in addition to
carbon, hydrogen, and oxygen. The great number of these carbon
compounds, many of which have been prepared by synthesis from the
elements, makes it necessary to consider them in a special branch of the
science known as organic chemistry.

The fact that carbon forms such a large number of compounds
is due to the facility with which its atoms, unlike those of other

658

CH. xxxin

CARBON AND THE HYDROCARBONS

659

elements, combine to form chains, which may have branches
(aliphatic, or fatty, compounds) ; or rings (cyclic, or aromatic, com-
pounds). Examples of such compounds are :

H H H H H H

I I I I I I
Hexane : H— C— C— C— C— C— C— H, contained in petrol

b.-pt. 69°. (aliphatic).

H H H H H H

Benzene :

b.-pt. 804°.

H\c/\c/H

, contained in coal tar (cyclic).

By means of the JT-ray method (p. 1018), the diameter of the
benzene ring has been estimated at 124 x 10 ~8 cm. ; its thickness
at 1-19 X 10~8 cm.

Allotropie forms of carbon. — Carbon is one of the most striking
examples of allotropy. The majority of organic compounds, when
heated without access of air, blacken, or char, evolve steam and
various volatile organic compounds, and usually inflammable
gases (e.g., methane, CH4), leaving finally a black residue of charcoal,
which, if a compound free from mineral matter and containing only
carbon, hydrogen, and oxygen (e.g., sugar, or cellulose) is used,
consists almost solely of carbon. The smoke produced on burning
oils with an insufficient supply of air also consists mainly of par-
ticles of carbon. That charcoal should be, chemically, the same
substance as the diamond would appear highly improbable to one
unacquainted with the fact ; its analogy with graphite, or black-
lead, would seem clearer, by reason of the colour, yet it is curious
that the composition of the diamond was elucidated (1775) before
that of graphite (1800). The identity of the three forms of carbon
was established by showing that equal weights of the pure sub-
stances, when burnt in oxygen, yield identical weights of carbon
dioxide, no other substance being produced. The amounts of heat
liberated in the three cases, however, are different : for 12 gm. of
carbon they are :

graphite : 94-26. kgm. cal.
diamond : 9443 kgm. cal.
charcoal : 97 -65 kgm. cal.

u u 2

660

INORGANIC CHEMISTRY

These differences are supposed to be due to different modes of
linkage of the carbon atoms in the molecules of the substances. It
is further assumed, on account of the extremely high temperature
at which carbon volatilises (the boiling point has been given as
3600°), and from other considerations, that the molecules are highly
polymerised, (C)n. Recent experiments, however (p. 1019), indicate
that the diamond consists of an assemblage of single atoms of
carbon, united one to the other by four valencies in space.

The diamond. — This mineral, which in its transparent varieties
forms the most beautiful and costly gem, has (been known from
very early times. It is found, as yellow rounded " pebbles," in
India, Brazil, New South Wales, Arkansas, and particularly at
Kimberley, in British South Africa. Most diamonds are small,
but the Cullinan diamond, discovered at Kimberley in 1905, weighed
about 1J lb., viz., 3025f carats (1 carat = 0-2054 gm.) ; this is
the largest yet discovered, and was cut into two brilliants of 516
and 309 carats.

Large colourless diamonds are the Pitt diamond (136-25 carats),
and the Koh-i-noor, originally 186 carats, but reduced to 106 by re-
cutting. The Hope diamond, 44-5 carats, is a fine blue stone, valued
at £25,000. The cause of the colour of diamonds is not clear : exposure
to cathode rays deepens the colour, which is lost on heating to 300-400°.

Black diamonds, known as carbonado, and bort, are of no value
as gems, but are very hard, and are used for rock-drills, and, when

crushed, for cutting and polishing clear
diamonds. The latter are pressed
against a revolving metal disc, covered
with diamond powder and oil.

The diamond crystallises in the
regular (or cubic) system ; forms
related to the cube or the octahedron,
sometimes with curved faces, pre-
dominate (Fig. 326). The curved
faces appear to have been formed by
the action of a solvent. By cutting,
however, the natural crystalline form
is obliterated, and an artificial shape,
which gives rise to a large amount of

internal reflexion, producing the " fire " of the stone, is impressed
upon it. The " brilliant," for example, consists of one larger
flat face, forming the base of a many-sided pyramid (Fig. 327).

Indian diamonds occur in river gravels and alluvial deposits, and are
separated by washing. They appear to have been transported by water.
At Kimberley the diamonds occur in situ in the original rock (" blue-

FIG. 326.— Diamond Crystal.

xxxm CARBON AND THE HYDROCARBONS 661

ground "), which is a weathered form of olivine, and runs in large
"pipes " downwards through the earth, cutting through strata of sand,
rock, and quartz. Masses of this earth are blasted out and allowed to
weather, when they crumble to light earth and a small quantity of
heavier mineral, consisting of pyrites, calcite, tourmaline, garnets, ekla-
dite, and, possibly, diamonds. The light material is washed off, and
the heavier residue carried b£ water over a bed of grease : to this the
diamonds adhere. The yield is variable ; in the richest mines it is
about 0-1 gm. per ton of earth.

The diamond is extremely hard, although fairly brittle : it is
scratched by no other substance (except possibly boron carbide,
B6C), and stands highest in Moh's scale of hardness, which com-
prises the following minerals : —

1. Talc. 3. Calcite. 5. Apatite. 7. Quartz. 9. Corundum.

2. Gypsum. 4. Fluorite. 6. Orthoclase. 8. Topaz. 10. Diamond.

Each mineral in the scale is scratched by all those below it. In
reality, the diamond is about 140 times harder than corundum.

The diamond has a high refrac-
tive index (2417 for the D-line),
a density of 3-0-3-52, and a high
dispersive power, exhibiting a
play of colours in white light. It
is transparent to JC-rays, whilst
all imitations are opaque. Dia-
monds are Used for cutting glass ; Fia 327 —Diamond cut as " Brilliant."
for this purpose a chisel-shaped

crystal-edge is necessary, since a splinter merely scratches glass
without cutting it.

The diamond resists the action of almost all chemical reagents ;
a mixture of potassium dichromate and sulphuric acid oxidises it
slowly at 200° to carbon dioxide. When strongly heated in the arc,
with exclusion of air, it is only superficially transformed into
graphite, which is the stable form at high temperatures, and is
produced from diamond and amorphous carbon alike. If heated
to 700-900° in air or oxygen, the diamond burns, leaving only a
trace of ash (0-02— 0-05 per cent., chiefly silica and oxide of iron) ;
bort may leave as much as 4 per cent, of ash. Diamonds are
attacked by fused sodium carbonate.

The combustibility of the diamond was foreshadowed by Newton,
who, arguing from the similarity of its refractive index to those of oil of
turpentine, camphor, and amber, suggested that it might be " an
unctuous [oily] substance coagulated." The Florentine Academicians
in 1694 heated a diamond in the focus of a powerful burning-glass : io

662

INORGANIC CHEMISTRY

CHAP.

glowed like a red-hot coal and disappeared. D'Arcet (1766) found that,
when a diamond was strongly heated in a closed crucible, it remained
unchanged. Davy and Faraday in 1813, using the original Florentine
lens, burnt a diamond in oxygen. It took fire, and continued to burn,
even if removed from the focus, with a steady brilliant light. Nothing
was produced but carbon dioxide, which rendered lime-water milky.
Tennant (1797) was able to burn diamonds by strongly heating them
with fused nitre in a gold tube : he found that as much carbon dioxide
was formed as Lavoisier had obtained from an equal weight of charcoal.

EXPT. 264. — The combustion of the diamond in oxygen may be
exhibited by heating a splinter of carbonado to whiteness by an electric

current in a spiral of fine platinum wire
supported by copper leads inside a jar of
oxygen (Fig. 328). A little lime-water
is shaken up with the gas afterwards.

•After many unsuccessful attempts to
prepare diamonds artificially, the pro-
blem was to some extent solved by
Moissan in 1893. He heated charcoal
with iron in the electric furnace to a
very high temperature. Fused iron
dissolves carbon (p. 982) ; on cooling
the iron slowly most of the carbon
deposits in the form of scales of
graphite, which are seen in a broken
piece of grey cast-iron. When the iron
is rapidly quenched, under ordinary
conditions, the carbon remains in solid
solution as the carbide, Fe3C, and
white cast-iron is produced. Moissan.
cooled the iron containing carbon
suddenly from 3500° to 350° by
taken from the electric furnace, into
The outer portion solidified at once,
and the still liquid portion imprisoned within it solidified in due
course. On dissolving away the iron with hydrochloric acid,
a residue was left containing three varieties of carbon : (1) a small
amount of graphite ; (2) curious brown twisted threads, apparently
formed under great pressure ; and (3) a denser portion which con-
tained microscopic diamonds, some black and some transparent.
It was usually considered that the important condition in Moissan ?s
experiment was the enormous pressure developed by the solidifica-
tion of the molten cast-iron inside the rigid outer skin which was
first formed, but Sir C. L. Parsons (1918) believes, from numerous

FlG. 328. — Combustion of the
Diamond in Oxygen.

plunging the crucible,
water or molten lead.

xxxin CARBON AND THE HYDROCARBONS 663

experiments, that the function of this skin is to prevent the escape
of occluded gases such as carbon monoxide, the presence of which
is essential to the formation of diamonds.

The presence of oxide of iron in diamond -bearing earth suggests that a
process similar to that used by Moissan may have been responsible for
the origin of the natural diamonds. Small clear diamonds have been
found in meteorites, and diamonds may be of celestial origin : the iron
may, however, have come from the interior of the earth.

Graphite. — Prior to ]779, molybdenum sulphide (MoS2) and
graphite (C) were confused together under the name motybdcena,
or black-lead, since both were soft black minerals with a metallic
lustre, giving a streak on paper, similar to that produced by lead.
Scheele, in that year, found that the former mineral gave a peculiar
solid acid (molybdic acid, MoO3) when roasted in the air, evolving
sulphur dioxide ; the name molybdena was reserved for this mineral,
whilst the other was called graphite (Greek grapho — I write),
plumbago, or black-lead, and considered to be a carbide of iron,
since it usually left a residue of oxide of iron when burnt, car-
bon dioxide being
formed. Scheele no-
ticed that graphite
deposits from
molten iron in

blast furnaces. This FlG> 329.— Manufacture of Graphite in the Electric Furnace.

variety is called kish.

In 1800, however, Mackenzie burnt graphite in oxygen and found
that it yielded almost as much carbon dioxide as an equal weight
of pure charcoal. The idea that it contained iron was not definitely
given up until perfectly pure graphite was first prepared by Brodie
in 1859, after which it was recognised as merely an allotropic form
of carbon.

Graphite is found in Borrowdale (Cumberland), Siberia, Ceylon,
India, and Bohemia ; enormous beds, 20-30 ft. thick, of nearly pure
graphite are found in the Eureka Black Lead Mines, California. It is
supposed to be of organic origin (see Coal). About 80,000 tons are
mined annually. Ceylon and Siberia supply most of the European
graphite.

Graphite is produced artificially on the large scale by the Acheson
process at Niagara : 2,500 tons were made in 1915. A mixture of
sand and powdered anthracite or coke (petroleum coke is best) is
heated very strongly for twenty-four to thirty hours by an electric
current. Carbon rods lead the current through the mass, which
is supported in a brick furnace and covered with sand (Fig. 329).

664 INORGANIC CHEMISTRY CHAP.

Apparently silicon carbide (carborundum) is first formed, and then
decomposed at the very high temperature, the silicon being volati-
lised : (1) Si02 + 30 - SiC + 2CO : (2) SiC =•• Si + C (graphite).

The product is very pure and soft, and free from grit. If treated with
water containing tannin, it forms a colloidal suspension, used as a lubri-
cant under the name of deflocculated graphite, or " aquadag " : when
kneaded with oil, the water is squeezed out and the suspension of
graphite in oil is called " oildag " (" dag " = deflocculated Acheson
graphite).

Graphite crystallises in grey, shining hexagonal plates, belonging
to the monoclinic system, which when rubbed flake off in thin layers ;
hence it has a greasy feel, makes a streak on paper, and acts as a
lubricant. It is also used (as " black-lead ") in polishing iron work
and granular gunpo\vder. An amorphous variety exists. Graphite has
a specific gravity of 2-1-2-6, and is a good conductor of heat and elec-
tricity : on account of the latter property it is used in the cores of
arc-carbons (p. 684), as anodes for electrolytic cells, and for covering
plaster moulds on which copper is deposited by the electrotyping
process (p. 809). Graphite burns only at a high temperature (about
690° in air), and, on account of its refractory character, is used for
making plumbago crucibles : these consist of 75 parts of plastic
clay, 25 parts of sand, and 100 parts of graphite, moulded and
baked. A granular mixture of graphite, carborundum, and clay is
used as a resistance in electric furnaces under the name of kryptol.
Mixed with a little plastic clay, and squirted into threads, graphite
is used in the manufacture of black-lead pencils.

Graphite is not attacked by dilute acids, or fused alkalies, or
when heated in chlorine. A mixture of potassium dichromate and
sulphuric acid slowly oxidises it to carbon dioxide. When moistened
with concentrated nitric acid and then heated, some varieties of
graphite (Borrowdale and Austrian) swell up : others (Ceylon and
American) do not. This is known as Luzrs test (1891).

The action of concentrated nitric acid on graphite is peculiar ;
whereas the diamond is not attacked by this reagent, and amor-
phous charcoal is oxidised to dark brown soluble substances con-
taining mellitic acid, C6(C02H)6 (Hatchett, 1805), and ultimately
to carbon dioxide, graphite is converted into a peculiar green or
yellow, almost insoluble, acid substance, known as graphitic acid
(Brodie, 1859). A mixture of nitric acid, potassium chlorate, and
sulphuric acid is usually employed as an oxidising agent.

Graphitic acid is very sparingly soluble in pure water, and reddens
moist litmus paper : it is microscopically crystalline or amorphous,
and has the formula C11H4O5. On heating it swells up and leaves
a fine black powder of pyre-graphitic oxide, C22H2O4. When treated
with hydriodic acid, graphitic acid takes up" hydrogen, forming

xxxm CARBON AND THE HYDROCARBONS 665

hydrographitie acid, which does not yield pyrographitic acid on
heating.

A mixture of potassium chlorate and concentrated sulphuric acid
converts graphite into a black substance containing hydrogen,
oxygen, and sulphuric acid, called graphon sulphate by Brodie.
On heating, this swells up, evolves gas, and then falls to a fine powder
of pure graphite (sp. gr. 2-25). If this is thrown on water, the
impurities sink, and the pure graphite remains floating on the
surface.

Amorphous carbon. — The following varieties of amorphous carbon
are usually described :

1. Charcoal : from wood, sugar, etc. 2. Lampblack : soot,
acetylene black. 3. Animal charcoal : bone-charcoal, ivory black.
4. Coke (coal, anthracite, etc.). 5. Gas carbon. 6. Electrode
carbon : arc carbons, etc. (artificial).

They are all black and opaque, the density and hardness depending
largely on the temperature at which they were formed. The X-ray
spectra show that they are all really microcrystalline, with the
same arrangement of the atoms as in graphite.

Charcoal. — The black residue, rich in carbon, obtained by heating
vegetable substances, such as wood or sugar, with exclusion of air,
is known as charcoal. The purest variety is obtained by heating
recrystallised cane-sugar in a large covered crucible until gases
cease to be evolved ; the resulting charcoal is heated to 1000°
in a graphite tube in a current of chlorine to remove residual
hydrogen as hydrogen /chloride, after which it is washed and ignited
in hydrogen to remove chlorine. Charcoal so prepared has a
density of 1-8, and ignites in air at 450°. Pure amorphous carbon
is also produced, mixed with magnesia, by burning magnesium
in carbon dioxide : it is free from hydrogen.

The low ignition temperature of charcoal, as compared with the
other forms of carbon, is seen from Moissan's results :

Wood
Diamond. Graphite, charcoal.

Evolution of carbon dioxide begins . . . 720° 570° 200°
„ abundant 790° 600°

Burns with flame 800-850° 690° 345°

Wood charcoal is largely used as fuel in countries where wood is
abundant. It is prepared by the destructive distillation of wood,
i.e., the decomposition of the latter into volatile parts (gas, water,
acetic acid, acetone, and tar), and non-volatile charcoal.

Dry wood on heating to 220° becomes brown, at 280° deep brown,
at 310° brown and friable ; above 350% black charcoal is produced.
The destructive distillation of wood, with production of tar, acid, and

666 INORGANIC CHEMISTRY CHAP.

spirit, was examined by Glauber in the seventeenth century. The
percentage of carbon in the charcoal never exceeds 78 when heated to
redness under ordinary pressure. By heating above 1500°, the residual
hydrogen falls to 0-62 per cent.

The manufacture of charcoal is carried out in : (a) pits or heaps
(meiler), (b) closed ovens or retorts. The charring of wood in meiler,
ordinary charcoal burning, is very old. A rough central chimney
is built of turf, and billets of wood stacked round it in a conical pile,
the whole being covered in with turf (Fig. 330). A lighted faggot
is dropped down the chimney, to kindle the wood, which burns
slowly, just sufficient air being admitted through holes at the bottom.
A part of the wood burns, and the heat generated chars the rest.
After some days the luminous flame from the chimney is replaced
by a blue flame of carbon monoxide. All the air-holes are now

FIG. 330. — Charcoal " Meiler."

stopped up, and the charcoal allowed to cool. About 24 per cent,
of the weight of the wood is obtained as charcoal ; all the volatile
products are lost.

In the^ modern process, based on Glauber's work, the wood is
heated in externally fired ovens, or iron retorts, from which air is
excluded. The volatile .liquid products are collected, and the
inflammable gas is used for heating the retorts. The liquid dis-
tillate consists of (a) a watery portion, the pyroligneous acid, con-
taining water, acetic acid, methyl alcohol, and acetone, which are
extracted ; (b) tar, which is valuable (e.g., Stockholm tar, from pine-
wood). The yields, from 100 parts of dry wood, are, by weight :
charcoal 25, tar 10, pyroligneous acid 40, gas 25,

Properties of charcoal. — Wood charcoal is a black, amorphous,
friable material, retaining more or less the original shape of the
wood, but diminished in volume. Although the specific gravity of
air-free charcoal is 14-1-9, the mass is very porous, and floats on

xxxin CARBON AND THE HYDROCARBONS 667

water. If the air is removed by placing the charcoal in water in a
bottle connected with an air-pump, the charcoal gives out bubbles
and slowly sinks. Charcoal is very permanent on exposure to air
and moisture ; charred oak stakes, planted in the bed of the Thames
by the Britons to resist the advance of Julius Caesar, were found
nearly two thousand years later, in the exact
spot indicated by Tacitus, and still sound at heart.
In virtue of its great porosity, charcoal readily
absorbs (or adsorbs) gases (Scheele, and Fontana,
1777).

EXPT. 265. — If a piece of recently ignited wood-
charcoal is passed into a tube of ammonia gas standing
over mercury (Fig. 331), the gas is rapidly absorbed ;
the charcoal takes up about 90 times its volume of
ammonia gas.

A very active form of charcoal is prepared by FIG. 331.— Absorption
heating the shell of the cocoanut ; 1 volume of &£^nia Qas by
such charcoal, quenched under mercury, absorbs
the following volumes of different gases (reduced to S.T.P.) at
the ordinary temperature :

Ammonia 171-7 Hydrogen phosphide 69-1

Cyanogen 107-5 Carbon dioxide 67-7

Nitric oxide 86-3 Carbon monoxide 21-2

Ethylene 74-7 Oxygen 17-9

Nitrous oxide 70-5- Nitrogen 15

The preferential absorption of ethylene by charcoal is applied in its
extraction from coal gas. Vapours of volatile liquids are absorbed
even more readily than gases : the volumes of ammonia, carbon dioxide,
steam, and alcohol vapour absorbed at 126-5° are 21-9, 16-6, 43-8, and
110-8, respectively. Generally speaking, the absorption increases
the nearer the gas or vapour is to its point of liquefaction at the tem-
perature of the experiment, and this supports Faraday's theory (p. 198),
that the gas is held by the charcoal in a highly condensed, possibly
liquid, layer upon its surface. McBain finds that the amount of gas
taken up increases slowly with lapse of time, due to a slow penetration
of the condensed layer into the interior (p. 197). At low tem-
peratures the absorbed amount increases rapidly (Dewar, 1904) :

Gas H2 N2 02 A He

0° 4 15 18 12 2 volumes

-185° 35 155 230 175 15 „

In this way high vacua (p. 193) may be produced, and gases
separated from one another.

668 INORGANIC CHEMISTRY CHAP.

EXPT. 266. — The condensed layer of gas held by the charcoal is very
reactive (Stenhouse, 1855). Place a crucible containing powdered,
recently ignited, charcoal in a jar of sulphuretted hydrogen. After
it has become saturated with the gas, transfer it to a jar of oxygen.
Ignition occurs.

Chlorine absorbed by charcoal unites with hydrogen passed over
it in the dark ; carbon monoxide and chlorine, or sulphur dioxide and
chlorine, unite when passed over charcoal, which acts as a catalyst,
to form carbonyl chloride, COC12, and sulphuryl chloride, SO2C12.

Charcoal also takes up many substances, e.g., metallic salts, and
organic substances such as alkaloids (e.g., quinine), and colouring
matters, from solutions (Lowitz, 1790). It removes fusel oil (amyl
alcohol) from crude spirit.

EXPT. 267. — Boil solutions of litmus and indigo with finely-powdered
animal charcoal, and filter. The filtrates are colourless.

Animal charcoal. — This material, also known as bone-black, is
prepared by the destructive distillation of bones in iron retorts.
The volatile products are : (a) a watery liquid which, unlike that
from wood, is alkaline, and contains ammonia and nitrogenous
organic bases ; (b) gases, and (c) bone-oil or Dippd's oil (containing
pyridine, etc.). The residue in the retort is a black mass containing
about 10 percent, of amorphous carbon disseminated through a
very porous, substrate, consisting of 80 per cent, of calcium phos-
phate (p. 609), together with calcium carbonate, etc. If the phos-
phate and other salts are dissolved out by hydrochloric acid, the
charcoal remains as ivory black.

Animal charcoal is largely used to decolorise sugar syrup, a
process introduced by Derosnes in 1812. This is an adsorptive
action, and is carried out by allowing the syrup to trickle through
towers 25-30 ft. high filled with small pieces of bone-black. When
the latter has become inactive, it is revived by reheating in retorts.
Finally it is burnt, yielding bone-ash (p. 609). Blood charcoal is
used for the same purpose.

Lampblack. — When carbonaceous fuels such as coal, wax, oil,
and turpentine (but not charcoal) are burnt with a supply of air
insufficient for complete combustion, part of the carbon separates
in the form of particles, forming smoke, which settles out on solid
surfaces as soot. A fine variety of soot, called lampblack, is prepared
as a pigment by burning turpentine, tar, etc., in a limited supply of
air, and collecting the soot by deposition on coarse blankets, or by
electrostatic precipitation. In America, natural gas is burnt under a
cooled, rotating metal disc, from which the lampblack is removed by
scrapers.

A very fine variety of lampblack, for use as a pigment, is prepared

xxxm CARBON AND THE HYDROCARBONS 669

by the spontaneous explosion of acetylene under 6 atm. pressure ;
pure hydrogen is produced at the same time : C2Ha = 20 -f- H2.

Lampblack contains up to 20 per cent, of oily impurities, which
may be removed by ignition in chlorine and hydrogen, as in the case
of sugar charcoal (p. 665) ; it is then a very pure form of carbon.
The density of lampblack is 1-78.

Goal. — The two varieties of amorphous carbon, coke and gas
carbon, are derived from coal, and since some varieties of coal
(anthracite) contain more than 90 per cent, of carbon, they will be
considered here.

Coal is a carbonaceous mineral, which is the final result of a series
of decompositions (which have occurred in the presence of a limited
supply of air) undergone by vegetable matter of the remote past.
High pressure, due to the weight of superimposed strata, wras pro-
bably also necessary in these changes. A portion of the carbon,
hydrogen, and oxygen was eliminated as carbon dioxide, water, and
methane (CH4), and the residue became increasingly rich in carbon.
The early stages of the decomposition of the vegetable matter were
probably caused by bacteria, and heating under pressure may have
played a part in the later stages. Distinct evidence of vegetable
remains in coal is disclosed by microscopic examination, and fossil
trees and plants are often found in the seams. The character of the
vegetable matter, and the manner in which it was covered by earthy
deposits, probably varied from case to case. Two theories have
been advanced to explain the origin of coal. Large beds of coal are
supposed to have been deposited in situ from vegetable remains ;
impure current -bedded local coal, such as cannel, is regarded as
derived from the burying of water-borne vegetable matter in a delta.

Stopes, from microscopic investigations, has recognised four
constituents in banded coal, viz., durain, fusain, vilrain, and clarain.
Although chemical methods have not given much useful information
as to these constituents of coal, their behaviour on coking, i.e., on
heating out of contact with air, has been shown by Lessing to be
different. Fusain yields a powdery coke ; in the case of durain the
coke is also very friable, whilst with clarain fusion and swelling
occur, with formation of a brown coherent coke. Vi train also
undergoes fusion, yielding a silver-white coke, which exhibits
excrescences.

Since the separation of the constituents of banded coal is a matter
of great difficulty, users of that fuel are more interested in the various
types of coal as they come from the mine ; although these may
be very heterogeneous, it is possible to give a broad general classi-
fication of coals, based on their behaviour during combustion or
gasification.

The first stage in the conversion of vegetable matter into coal
is represented by peat, which consists of accumulations of vegetable

670 INORGANIC CHEMISTRY CHAP.

matter, chiefly mosses and bog-plants, which have undergone partial
change, and still preserve evidences of organic structure, although the
deeper layers may be more compact and homogeneous. The next
stage is represented by lignite, or brown coal, which is more compact
than peat, and is lustrous, although impressions and remains of
vegetable fragments, leaves, etc., are still distinct and numerous.
Large beds of lignite occur, near the surface, in many parts of
Germany, Hungary, and the Mississippi Valley, and are utilised as a
cheap steam-raising fuel. Jet is a hard variety of lignite, used for
ornaments.

The next stage of the process leads to the very important types
of bituminous coal, i.e., common coal. These, as mentioned above,
are complex : distinct evidences of vegetable origin are still present,
and the original plants are sometimes found fossilised. Bituminous
coals burn with a bright smoky flame, and are further divided into
caking and non-caking coals, according as they do or do not soften
and fuse together on burning or coking. Cannel coal is a compact,
non-lustrous, variety, dull grey or black in colour, breaking with a
conchoidal fracture, and yielding a large amount of gas and little
coke. Splinters of cannel coal burn like candles when ignited,
hence the name.

The latest stages in coal-formation consist chiefly of carbon,
and are known as anthracite. Anthracite has a high ignition
point, usually a brilliant lustre, and a conchoidal fracture, and does
not burn with a flame. It is used in firing ships' boilers, since it gives
an intense heat on combustion. Anthracite occurs locally in many
coal-fields, such as South Wales, Scotland, and Pennsylvania.
Graphite may represent the ultimate stage of the decomposition,
since it always contains a little hydrogen.

TABLE OF ANALYSES OF COALS.

Mois-
C H O N S Ash. ture. Coke.

1. Air-dried wood 42 5 37 1 15 25

charcoal

2. Air-dried peat 57-03 5-63 29-67 2-09 5-58

3. Lignite ... 44-93 3-12 12-51 0-64 0-50 4-43 34-28

Coking coals : —

4. Northumberland 81-41 5-83 7-90 2-05 0-74 2-07 1-35 66-70

5. Wales ... 83-78 4-79 4-15 0-98 1-43 4-91 — 72-60

6. Staffordshire ... 78-57 5-29 12-88 1-84 0-39 1-03 11-29 57-21

7. Wigan Cannel ... 80-07 5-53 8-08 2-12 1-50 2-70 0-91

Anthracites :—

8. South Wales ... 90-39 3-28 2-98 0-83 0-91 1-61 2-00

9. Pennsylvania ... 92-59 2-63 1-61 0-92 — 2-25 —

xxxm CARBON AND THE HYDROCARBONS 671

The total output of coal amounts to about 1000 million tons per
annum. The annual outputs in Great Britain have been, in millions
of tons : 1913, 287 ; 1916, 257 ; 1917, 249 ; 1918, 214-217., The
diminishing production is a most disquieting and serious fact.

The calorific power of a fuel is expressed as the number of British
thermal units (B.Th.U., i.e., the number of Ib. of water raised 1°F.
in temperature), evolved by the complete combustion of 1 Ib. of the
fuel, the water formed being supposed condensed to the liquid
state. The following are examples : peat (30 per cent, moisture),
1462 ; lignite, 7065 ; bituminous coal, 15,000 ; anthracite, 15,300.

Carbides. — Compounds of metals with carbon are called carbides.
Of the alkali metals, only lithium combines directly with carbon,
forming Li2C2. Calcium is the only metal of the alkaline earths
which unites directly with carbon, forming CaC2 ; carbides of all the
metals of this group are, however, produced by heating the oxides
with carbon in the electric furnace : MO +30 = MC2 -f CO.
Beryllium is the only metal of the sub-group II which combines
directly with carbon, forming BeC?. Of the earth metals, aluminium
alone unites with carbon to form A14C3 ; the rest form carbides when
their oxides are strongly heated with carbon. Iron, chromium,
tungsten, and molybdenum form carbides directly, which are not
attacked by water (Fe3C, Cr3C2, Cr4C, W2C, WC, MoC, Mo2C) ;
manganese and uranium form Mn3C and U2C3, which are
decomposed by water. The remaining metals dissolve carbon but
do not form carbides.

By the action of water on carbides, hydrocarbons, i.e., compounds
of carbon and hydrogen, are produced. Alkali and alkaline-earth
carbides form acetylene C2H2:CaC2 + 2H20 = Ca(OH)2 -f C2H2.
Beryllium and aluminium carbides give methane, CH4:A14C3 -f
12H2O = 4A1(OH)3 -f 3CH4. The carbides of the rare metals, e.g.,
thorium carbide, ThC2, and uranium carbide, U2C3, form gaseous,
liquid, and solid hydrocarbons ; manganese carbide evolves a
mixture of methane and hydrogen.

Petroleum. — Petroleum consists of liquid hydrocarbons, contain-
ing," in its natural state, dissolved gaseous and solid hydrocarbons.
It is purified by agitating with concentrated sulphuric acid, and then
washing with water, and is separated into fractions by distillation.
The portion coming over between 40° and 150° is petrol, and consists
chiefly of the hydrocarbons C6H14, C7H16, and C8H18. The distillate
between 150° and 300° is ordinary petroleum, or paraffin oil, used for
burning in lamps. The residue is vaseline. In some cases paraffin
wax is contained in the residue.

Since liquid hydrocarbons similar to petroleum are formed by the
action of water on metallic carbides, it has been suggested that this
reaction may account for the formation of petroleum in the interior
of the earth (Mendeleeff, 1877 ; Moissan, 1896). Another hypothesis

672 INORGANIC CHEMISTRY CHAP.

(Engler, 1888) is that petroleum has been formed by destructive
distillation of organic remains, particularly fish.

The hydrocarbons present in paraffin oil are very inert towards
chemical reagents (hence the name, from parum affinis). They are
called saturated hydrocarbons, since they do not form addition com-
pounds. The numerous members of the series have the general
formula CnH2n + 2, and are derived from the simplest, methane, CH4,
by successive addition of CH2. A series of compounds, the
successive members of which differ in composition by CH2, is called
a homologous series.

The cracking of oils. — The decomposition of hydrocarbons of high
boiling point to simpler hydrocarbons of relatively low boiling
point is effected by heating with exclusion of air, and is known as
" cracking." By this process, for example, it is possible to convert
heavy petroleum into petrol. A heavier residue, and gas, are at the
same time produced. A catalyst, such as nickel, or chromium
oxide, is used, and the reaction is carried out at 350-450°.
Unsaturated hydrocarbons are also formed.

HYDROCARBONS .

Methane, or marsh gas, CH4.— The first member of the paraffin
series of hydrocarbons is methane, or marsh gas, CH4, which is formed
by the bacterial decay of vegetation (cellulose) at the bottom of
marshy pools', and is liberated in bubbles when the mud is disturbed
with a stick. It also occurs occluded in coal, and escapes when the
pressure is relieved, forming the fire- damp of the mines, which, when
mixed with air, causes explosions on ignition. The gas often issues
in large quantities from " blowers." or fissures in the coal, and con-
tains 80-98 per cent, of methane, with some carbon dioxide and
nitrogen. Natural gas, from petroleum wells, contains more than 90
per cent, of methane, and is used for heating purposes instead of
coal.

Methane is formed by the direct union of carbon and hydrogen
on heating: C -f- 2H2±^: CH4. By circulating hydrogen over
heated sugar-charcoal more than 95 per cent, of the theoretical
yield is produced. Between 1100° and 2100°, at pressures up to
200 atm., methane is the only saturated hydrocarbon formed :
ethylene and acetylene are formed in smaller amounts. The per-
centages of methane in equilibrium with carbon and hydrogen
at atmospheric pressure are: 850°, 2-5; 1000°, 1-1; 1100°, 0-6.
Methane is produced when hydrogen mixed with carbon monoxide
or dioxide is passed over reduced nickel at 250° or 350°, respectively :

CO + 3H2 = CH4 -f H20.

CARBON AND THE HYDROCARBONS

673

In the laboratory, methane is usually prepared by heating a
mixture of fused sodium acetate with three times its weight of soda-
lime, in a hard glass or copper flask (Fig. 332) : it is collected over
water : CH3-COONa + NaOH = Na2C03 + CH4 (EXPT. 268).
Prepared in
this way, the
gas is not very
pure : it may
contain up to
8 per cent, of
hydrogen, and
also some
ethylene,
C2H4, which
causes it to
burn with a
slightly lumi-
nous flame.

Pure me-
thane is pre-
pared by the
action of

Water On zinc FIG. 332. — Preparation of Methane.

methyl, or on

an ethereal solution of magnesium methyl bromide obtained by
dissolving magnesium powder in a solution of methyl bromide in
dry ether :

Zn(CH3)2 + 2H20 = Zn(OH)2 -f 2CH4.
MgCH3Br + H20 = Mg(OH)Br + CH4.

The nearly pure gas produced by the action of water on aluminium
carbide: A14C3 + 12H2O = 4A1(OH)3 -f 3CH4, is purified from
hydrogen by adding a little more pure oxygen than is necessary to
combine with the hydrogen, and passing over palladium black.
The excess of oxygen is then removed by pyrogallol, and perfectly
pure methane is left.

Properties of methane. — Pure methane is a colourless, odourless
gas which is not poisonous. Methane is liquefied at 0° under a
pressure of 140 atm., b.-pt. — 164°, m.-pt. — 184°. The critical
temperature and pressure are — 82*85° and 45-6 atm. The relative
density of methane is 7-97 ; the theoretical value is 15-9 -r 2 = 7-95 ;
hence the gas is slightly more compressible than an ideal gas. It is
sparingly soluble in water : 100 vols. of water dissolve 5-5 vols. at
0°, and 3-5 vols. at 20° ; but is somewhat more soluble in alcohol.

Methane is decomposed by heat directly into carbon and hydrogen :
the decomposition is inappreciable at 700°, and sixty times faster at

x x

674 INORGANIC CHEMISTRY CHAP.

985° than at 785°. It burns in air, or oxygen, with a pale blue non-
luminous flame : CH4 -f- 202 = C02 + 2H2O ; its ignition point in air
is 650-750°. When mixed with oxygen or air, it forms a violently
explosive mixture : 1 vol. of methane requires 2 vols. of oxygen, or
9-5 vols. of air, for complete combustion. The lowest percentage of
methane in air necessary for the propagation of flame is 3*75—4
by volume : the lowest ignition temperature is stated to be 500°.
The composition of the gas is found by exploding a measured volume
with oxygen, and measuring the volume of the residual carbon
dioxide. If hydrogen is present it is first removed by adding
oxygen and passing over palladium-asbestos at 100°. Hydrogen
alone burns (fractional combustion).

By the slow combustion of methane, which occurs when a mixture
of the gas with air or oxygen is passed over heated porcelain, traces
of formaledhyde, H-COH, are formed : CH4 + 02 == H-COH + H20.

According to Bone, the combustion of methane arid of other hydro-
carbons occurs by the entrance of oxygen into the molecule, where it
is distributed between the carbon and hydrogen, giving unstable
hydroxylated molecules which undergo thermal decomposition into
simpler products ; these, in turn, may undergo oxidation or further
thermal decomposition :

0 CO -}- H2 -f H2O (thermal]

CH4 + 02->CH2(OH)2-> H-CHO + H2O + O

TreeTte111 fopmaldehyde ^ H.CO2H + IS.^ (oxidation]

formic acid

Dalton found (1805) that if methane is mixed with half its volume
of oxygen and fired, the mixture doubles in volume, with the for-
mation of carbon monoxide and hydrogen : 2CH4 -j- 02 = 2CO
+ 4H2. On adding a further 4 vols. of oxygen, the gas may again
be fired : 2CO + 4H2 + 4O2 = 2CO2 + 4H2O.

If 1 vol. of methane is mixed with 2 vols. of chlorine in a cylinder,
and the mixture ignited by a taper, it burns with a flame, producing
fumes of hydrochloric acid and a black cloud of carbon : CH4 -f
2C12 = 4HC1 -f C. A mixture of equal volumes of chlorine and
methane, on exposure to diffuse daylight, slowly reacts with the

S reduction of hydrogen chloride and methyl chloride : CH4 -f- C12 =
H3C1 + HC1. With excess of chlorine, hydrogen is progressively
replaced by chlorine until carbon tetrachloride. CC14, is formed as a
final product :

1. CH4 ,+ C12 == HC1 + GH3C1, methyl chloride.

2. ^Sfi -f C12 = HC1 + CH2C12, methylene chloride.

3. CH2C12 + C12 = HC1 + CHC13, chloroform.

4. CHC13 + C12 = HC1 -f CC14, carbon tetrachloride.

These are examples of substitution ; 1 atom of hydrogen is dis-

XXXIII

CARBON AND THE HYDROCARBONS

675

placed from the molecule and replaced by an atom of chlorine.
The atom of hydrogen displaced combines with the second atom of the
chlorine molecule to form a molecule of hydrogen chloride. Since
methane can react only by substitution, or decomposition, not by
addition, it is called a saturated hydrocarbon.

Ethylene, C2H4. — By the interaction of hydrogen and carbon at
high temperatures, besides methane, traces of ethylene, C2H4, are
formed, which may be absorbed by passing the cooled gas over
charcoal cooled in liquid air. Most of the ethylene, however, is
decomposed at the high temperature. At 1200° the ratio of methane
to ethylene is 100 : 1 ; at 1400° it is 10 : 1.

Ethylene is prepared by dehydrating alcohol by means of zinc
chloride, boron trioxide, phosphorus pentoxide, concentrated

FIG. 333.— Preparation of Ethylene.

sulphuric acid, or syrupy phosphoric acid : C2H5-OH = C2H4 -f
H20. With sulphuric acid, ethylsulphuric acid, C2H5-HSO4, is
first formed and then decomposed : (1) C2H5-OH -f- H2SO4 =
C2H5-HS04 + H20. (2) C2H5-HS04 = H2SO4 + C2H4. This
method of preparation appears to have been discovered by Becher.

EXPT. 269. — Twenty-five gm. of alcohol and 150 gm. of concentrated
sulphuric acid are heated in a 2-3 litre flask at 160-170°, and a mixture
of 1 part of alcohol and 2 parts of sulphuric acid dropped in from a tap-
funnel. The gas is washed with sulphuric acid to remove alcohol and
ether vapour, and with caustic soda to take out sulphur dioxide. The
ethylene is collected over water (Fig. 333).

x x 2

676 INORGANIC CHEMISTRY CHAP.

EXPT. 270. — According to Newth's method (1901), alcohol is dropped
by a tube reaching to the bottom of a distilling flask into 50 c.c. of
syrupy phosphoric acid, which has been boiled till the temperature
rises to 200-220° ; or alcohol vapour from one flask passed through
the phosphoric acid at 220° in a second flask. The gas is passed through
a tube cooled in ice. This gives a very pure gas, and is probably the
best method of preparation.

Ethylene is formed from a mixture of carbon monoxide and
hydrogen in contact with heated finely-divided nickel : 2CO -{- 4H0
= 2H20 + C2H4.

Properties of ethylene.— Ethylene is a colourless gas with a
peculiar sweet smell. It is slightly soluble in water, and very soluble
in alcohol. B.-pt. — 104-3°, m.-pt. — 169°; critical temperature
9-5°, critical pressure 51 atm. On sparking the gas is decomposed
into carbon and hydrogen. When passed through a red-hot tube
it gives hydrogen, acetylene, and methane, with deposition of
a brilliant film of amorphous carbon.

According to Bone and Coward, the thermal decomposition may be
represented by the following scheme :

* (a) C2H2 + H2.

H2C:CH2 -> 2CH) + H2— >(&) 2C + H2 + H2.

\ (c) C2H2 + 3H2 = 2CH4.

The radical CH • is supposed to have a transient existence : it may
undergo polymerisation, with formation of complex ring compounds
(c/. p. 680).

Ethylene burns in air with a smoky, luminous flame : in oxygen
the flame is very bright, and does not smoke. When mixed with
oxygen in the proportions of 1 : 3 by volume and ignited, ethylene
explodes violently, and undergoes complete combustion :
C2H4 + 3O2 = 2C02 + 2H2O. If mixed with an equal volume of
oxygen and fired by a spark, an expansion occurs, and carbon mon-
oxide and hydrogen are formed : C2H4 -f- 02 = 2CO -f- 2H2. If
the resulting mixture, which burns with a blue flame in air, is mixed
with half its bulk of oxygen and again exploded, carbon dioxide and
steam are formed : 2CO + 2H2 -f 202 = 2C02 + 2H20.

The combustion of ethylene is represented in Bone's scheme as
follows :
H-C-H+o2H-C-OH_H2oH-COH+o2 H-COOH+o2 HO-CO-OH-2H2oCO2

II -» I! -> -» -> ->

HOH H-OOH H-COH H-COOH HO-CO-OH CO2

(hypothetical) 2 formaldehyde 2 formic acid 2 carbonic acid

(hypothetical)

xxxin CARBON AND THE HYDROCARBONS 677

Hydrogen and carbon monoxide arise from the thermal decompo-
sition of formaldehyde, as in the oxidation of methane. The liberation
of free carbon in the flame is not included in this hypothetical scheme.

If ethylene is mixed over water with an equal volume of chlorine
and the mixture exposed to light, contraction occurs and oily drops
collect on the surface of the water. These consist of ethylene
dichloride, C2H4C12, or Dutch liquid, formed by the direct addition
of chlorine to the double bond in the ethylene molecule :

H2C:CH2( -> H2-OCH2) -f C12 = CH2C1-CH2C1.

On account of this reaction, ethylene was first called olefiant gas
(i.e., oil -forming gas) by Fourcroy. Ethylene dichloride was
discovered by the Dutch chemists, Deimann and Paets van Troost-
wyck, in 1781. If passed into bromine covered with a layer of
water, ethylene combines with the halogen to form a colourless,
pleasant-smelling liquid, ethylene dibromide, C2H4Br2, or CH2BrCH2Br,
similar to the dichloride.

A mixture of 1 vol. of ethylene and 2 vols. of chlorine, when ignited,
burns with a red flame, fumes of hydrochloric acid and a dense black
cloud of soot being formed : C2H4 -f 2C12 = 4HC1 + 20.

Ethylene forms additive compounds with iodine, hydrobromic acid,
and hydriodic acid at 100°, but not with hydrochloric acid :
CH2:CH2 + HBr = CH3-CH2Br. When mixed with hydrogen and
passed over reduced nickel at 130-150°, it forms the saturated hydro-
carbon ethane : C2H4 + H2 = C2H6, or CH2:CH2 + H2 = CH3-CH3.
Hypochlorous acid forms glycol chlorohydrin : CH2:CH2 + HOC1 =
CH2-OH'CH2C1. Cold dilute potassium permanganate solution is
decolorised by ethylene, hydrated manganese dioxide is deposited,

CH2-OH
and the ethylene is oxidised to glycol : CH2 : CH2 -f- H2O -j- O = |

CH2-OH.

This reaction with potassium permanganate is characteristic of a
double bond between carbon atoms :

C:C <T + H2O + O = -COH-COH-

Concentrated sulphuric acid absorbs ethylene, slowly on shaking
at the ordinary temperature, rapidly at 160-170°, with the formation
of ethylsulphuric acid, or sulphovinic acid, C2H5-HS04: C2H4 -f
H-HS04= C2H5-HS04. When this is boiled with water, alcohol
is produced : C2HB-HSO4 + HOH = C2H6-OH + H2SO4. Fuming
sulphuric acid rapidly absorbs ethylene, a reaction used in gas
analysis as an alternative to absorption by bromine water, for the
estimation of ethylene. Ethionic acid, C2H4-H2S2O7, and carbyl
sulphate, C2H4S206, are formed.

Acetylene, CM2- — By the action of water on the carbide of potass-

678

INORGANIC CHEMISTRY

CHAP.

ium formed in the preparation of the metal from potassium car-
bonate and charcoal, Edmund Davy (1836) obtained a new hydro-
carbon, which was rediscovered by Berthelot in 1859, and called by
him acetylene. He showed that it is formed when ethylene or
alcohol vapour is passed through a red-Jiot tube, but an important
fact discovered in this work was the formation of acetylene by

direct synthesis from
its elements, when
an electric arc burns
between carbon
poles in an atmo-
sphere of hydro-
gen (Fig._^ 334) :
2C -}- H2 * — C2H2.
are also formed,

C.H,

Fio. 334.— Berthelot's Synthesis of Acetylene.
Small quantities of methane and

ethane
apparently by independent reactions.

Acetylene is produced when a Bunsen burner " strikes back,"
i.e., when the coal gas burns at the lower small jet, with a limited
supply of air and in contact with the metal tube, which cools the
flame. The peculiar smell noticed is usually said to be due to the
presence of acetylene ; although this odour always accompanies
the formation of acetylene in the reaction, it appears to be due to
some other substance. The
acetylene probably arises from
the thermal decomposition of the
ethylene in the coal gas.

EXPT. 271. — The presence of
acetylene in the gas issuing from
the burner is readily detected by
holding over it a large globe wetted
inside with an ammoniacal solution
of cuprous chloride (p. 816). The
dark blue liquid rapidly becomes
covered with a red film, owing
to the precipitation of cuprous
acetylide, Cu2C2, an explosive
substance.

Acetylene is prepared for use in illumination by the action of water
on calcium carbide : CaC2 -f 2H2O = Ca(OH)2 + C2H2.

EXPT. 272. — Cover the bottom of a conical flask with a layer of sand,
and place on this a small heap of granular calcium carbide. Fit the
flask with a rubber stopper carrying a dropping funnel, and inlet and
outlet tubes for gas (Fig. 335). Displace the air with a current of
coal gas, and then allow water to drop slowly on the carbide. Acetylene

FIG. 335. — Preparation of Acetylene.

xxxin CARBON AND THE HYDROCARBONS 079

is rapidly evolved, and will burn at the end of the exit tube with a very
luminous, smoky flame. The acetylene prepared from commercial
carbide has an unpleasant smell, due to the presence of impurities, such
as phosphinc, PH.,. These may be removed by passing through a
solution of bleaching powder.

Acetylene generators, used for the preparation of the gas, act
either on the principle of the Kipp's apparatus (p. 185), or else a
regulated stream of water is allowed to drop on the carbide.

Acetylene is formed when ethylene dibromide is dropped into boiling
alcoholic potash. The bromine is removed together with hydrogen,
in the form of two molecules of hydrobromic acid : CH2Br-CH2Br =
C-HCH + 2HBr. The compound C2H3Br is formed in an inter-
mediate stage.

Properties of acetylene. — Acetylene is a colourless gas which is
said to have an ethereal smell when pure, but ordinarily has an
unpleasant odour. When strongly cooled it forms a white solid,
subliming at — 85°. Under T25 atm. pressure the solid melts
at — 81° to a colourless liquid. The critical temperature of
acetylene is 35-5°; the critical pressure is 61-5 atm. The gas
dissolves in its own volume of water, and is very soluble in alcohol.
Acetylene ignites at 480° in air, burning with a very smoky, luminous
flame, but if it is supplied to special burners under a pressure of
2-8 in. of water, so as to escape through fine capillaries and mix
with a regulated amount of air, the flame is very bright and does not
smoke. Acetylene 4 explodes with oxygen with extreme violence:
it is unsafe to try* the experiment with ordinary precautions, as
strong glass vessels are shattered by the explosion.

Mixtures of acetylene and air, in proportions varying from 4 : 5
to 4 : 80, are explosive. Coal gas is only explosive when mixed
with air within the limits 1 of gas to 5-13 of air, and the lower limit
of explosion for methane is 5'3 per cent, in air. The danger of
explosion from escape of acetylene is therefore much greater than
with coal gas.

Acetylene is less poisonous than carbon monoxide, or even than
coal gas (which contains the latter) ; it forms with the haemoglobin
of the blood a compound which, unlike that produced by carbon
monoxide, is unstable, and is readily decomposed by aeration.

Acetylene is formed from its elements with considerable absorption
of heat : 2C -f H2 = C2H2 — 47-8 kgm. cal. It is for this reason
unstable, and readily explodes under moderate pressure. The gas
is therefore generated only as required, or is absorbed in acetone,
which dissolves 300 vols. of the gas under 12 atm. pressure. The
acetone is soaked up in porous material (" kapok "), contained in a
steel bottle (p. 189). The chief use of acetylene is for illumination,
and for the oxy-acetylene blowpipe.

680 INORGANIC CHEMISTRY CHAP.

The unsaturated character of acetylene is shown by its capacity of
forming addition compounds. Chlorine combines violently with the
gas, forming the dichloride, CHC1:CHC1, and the tetrachloride,
CHC12'CHC12. Under the influence of platinum black, acetylene
combines with two or four atoms of hydrogen, forming ethylene or
ethane, C2H4 or C2H6, respectively. Hydrobromic acid forms
CH2:CHBr, and CH3-CHBr2 (ethylidene bromide, isomeric with
ethylene dibromide, CH2BrCH2Br).

If acetylene is passed into a boiling solution of 3 vols. of sulphuric
acid and 7 vols. of water to which a few per cent, of mercuric sulphate
is added, acetaldehyde, CH3-CHO, is continuously formed and distils
off. The first reaction is the formation of a mercury compound
which is decomposed by the acid. With mercuric chloride solution,
a white precipitate of the compound, trichloromercuriacetaldehyde,
(ClHg)3OCHO, is formed, which is hydrolysed by acids to acetaldehyde.
Another compound, HgCl2,C2H2, is also formed in alcoholic solution.
The addition of water to acetylene, with formation of acetaldehyde,
which occurs in the reactions :

2C2H2 + 2H2O + 6HgCl2 = 2(C]Hg)3OCHO + 6HC1 ;
2(ClHg)3OCHO + 6HC1 = 2CH3-CHO + 6HgCl2,

is the basis of an important technical process. From acetaldehyde,
by reduction with hydrogen, alcohol, CH3'CH2-OH, can be obtained ;
on oxidation, aldehyde yields acetic acid, CH3-COOH ; both are
important materials. The use of alcohol, alone or mixed with benzene,
as a fuel in motor engines instead of petrol, is now an accomplished
fact, and the cheap production of calcium carbide is also possible
where power from water is available. It is not certain, however, that
it may not be cheaper to obtain alcohol by fermentation.

When acetylene is heated to dull redness, a complicated polymerisa-
tion reaction occurs, and a liquid mixture of hydrocarbons is obtained,
one of which is benzene : 3C2H2 = C6H6. This is an example of the
conversion of an aliphatic into an aromatic hydrocarbon. A certain
amount of the acetylene appears also to break up into the free radicals,
CH • , which decompose into carbon and hydrogen, the latter combining
with the CHj to form methane, CH4.

Coal gas. — The destructive distillation of coal, with the formation of
gas, was first carried out by the Rev. John Clayton in 1688, the results
being published in 1739. It was also described by Bishop Watson,
who found that a permanent gas, tar, and a watery liquid were
formed. The use of coal gas as an illuminant was introduced by
William Murdoch in 1792 ; in 1798 he installed a gas plant for
lighting the factory of Boulton and Watt, at Soho, near Birmingham.
Gas lighting was introduced into Manchester factories in 1808, the
first public gas-works being erected at Salford, and about the same
time gas lighting was used, on a very small scale, in London. The

XXXIII

CARBON AND THE HYDROCARBONS

681

capital was

lighted by gas
in 1812, Paris
following in
1815, but the
use of gas in
dwelling - houses
came much later.
In the gas-
works bitumin-
ous coal, alone
or mixed with
cannel, is heated
in fireclay retorts,
A (Fig. 336),
which may be
horizontal, in-
clined, or verti-
cal. These are
heated by pro-
duc er gas
(p. 705), formed
by passing air
and steam
through incan-
descent coke
resting on bars
in a firebrick
well beneath
each retort, the
gas formed being
burnt under the
latter. The gas
evolved from the
coal in the re-
torts passes, by
way of vertical
ascension pipes,
to a common
hydraulic main, B,
consisting of a
long horizontal
iron cylinder
into which these
pipes lead. The
main serves as

Q-rnn

682 INORGANIC CHEMISTRY CHAP.

a water-seal, preventing gas passing back when a retort is
opened.

In the hydraulic main separation occurs into crude gas, ammo-
niacal liquor, and tar. The gas leaving the main, at 50-60°, contains
the following impurities, which must be removed :

Ammonia, 0-7 - 1-4 per cent, by volume.

Hydrocyanic acid, 0«05 — 0-15 „ ,, ,,

Sulphuretted hydrogen, 0*9 — 1-7 ,, „ „

Carbon disulphide, 0-02 — 0-04 „ „ „

More tar is separated in the condensers, D, consisting of a series of
vertical iron cooling pipes. Ammoniacal liquor is deposited with
the tar, and the two collect in the tar-well, C. The gas next passes
to a special tar separator, and is drawn by an exhauster to the scrubbers,
E. These are iron towers packed with coke, down which water
passes. The rest of the ammonia is thus removed.

The gas passing from the scrubbers contains as impurities carbon
dioxide, some sulphuretted hydrogen (a portion of each gas is de-
posited with the ammonia in the previous cooling and scrubbing),
and carbon disulphide. A cubic foot of crude gas usually contains
800 grains of sulphur as H2S and 40 grains as CS2. These impurities
are removed by the purifiers, F, in which the gas passes over trays
covered with slaked lime, or hydrated ferric oxide. The lime
absorbs sulphuretted hydrogen, forming calcium hydrosulphide :
Ca(OH)2 + 2H2S = Ca(SH)2 + 2H20. The oxide of iron decom-
poses the sulphuretted hydrogen with formation of ferric sulphide,
Fe2S3, or a mixture of ferrous sulphide and sulphur : Fe203 -J-
3H2S = Fe2S3 + 3H20. The oxide is " revived " by exposure to
air, when sulphur is separated, and hydrated ferric oxide regenerated :
2Fe2S3 + 302 = 2Fe2O3 + 6S. When it contains 50 per cent, of
sulphur, the " spent oxide " is burnt to produce sulphuric acid
(p. 503). The old method of removing carbon disulphide was to
pass the gas through lime previously used to remove sulphuretted
hydrogen ("foul lime"), when a thiocarbonate is formed:
Ca(SH)2 -f CS2 = CaCS3 -f H2S. The sulphuretted hydrogen evolved
is removed in a second oxide purifier. Usually the carbon disul-
phide is left in the gas, or removed by a catalytic process, in which
the gas is passed over nickel at 450° : CS2 + 2H2 == 2H2S -f C.
The sulphuretted hydrogen is removed as usual.

The purified gas is now passed to the gas-holder, G, a counterpoised
iron bell sealed below by water. From this it is distributed to the
mains. The purified gas should contain less than 1 part of H2S
per 10,000,000 parts, i.e., it should not blacken lead acetate paper.

Cyanides, which are of value, are removed from the crude gas by
passing through ammoniacal liquor containing ammonium sulphide,

CARBON AND THE HYDROCARBONS

683

solution of ammonium
= (NH4)2S + NH4CNS.
gas, in percentages by

43 —

55 ^ " Diluents," non-illumi-

35 > nating, but

heat-

1 1 . producing.

2-5 —

5 Hluminants.

2 —

12 \

0 —

3 > Impurities.

0 —

1-5/

XXXIII

with powdered sulphur in suspension. A
thiocyanate is formed : (NH4)2S2 + NH4CN

The average composition of genuine coal
volume, is as follows :

Hydrogen,

Methane,

Carbon monoxide,

Olefines, acetylene and benzene

Nitrogen (from air leakage)

Carbon dioxide

Oxygen

The calorific power of good coal gas is about 16,000 B.Th.U. per Ib.
or 600 B.Th.U. per cu. ft.

The hydrogen is derived from the thermal decomposition, at
700-800°, of gaseous hydrocarbons in contact
with the hot walls of the retort. The carbon
formed is deposited as a hard greyish -black mass
of gas carbon, which is removed by chipping.
This is a very pure form of amorphous carbon,
of density 2-35, which is a good conductor of
electricity, and is used for the pencils of arc lamps
or in electric batteries.

The luminosity of genuine coal gas flames is
due entirely to the 5 per cent, of olefine hydro-
carbons, acetylene (0-06-0-07 per cent.), and
benzene vapour.

EXPT. 273. — The effect of such hydrocarbons on
the luminosity of flames may be illustrated by fitting
a hard glass jet to each arm of a Y-tube, in one arm
of which is a piece of cotton- wool soaked in benzene
(Fig. 337), attaching the tube to a hydrogen apparatus, and lighting
the two jets. The hydrogen saturated with benzene vapour burns with
a luminous flame.

Modern coal gas is usually mixed with water gas (p. 705), produced
by blowing steam over the red-hot coke in the retorts : C rf H2O ^
CO + H2. In this way the percentage of carbon monoxide is increased,
and that of methane diminished. Owing to leakage of air, the nitrogen
content often exceeds 20 per cent. The quality of the resulting gas
is consequently mediocre as compared with the old genuine coal gas,
or coke-oven gas, and this has led to a very natural prejudice against
gas for domestic fires.

Coke. — The red-hot residue in the gas retorts is raked out, or
pushed out by rams, through doors opened at the front and back,

FIG. 337. — Lum-
inosity imparted
to Hydrogen
Flame by Benz-
ene Vapour.

684

INORGANIC CHEMISTRY

CHAP.

and is quenched with water. It is known as gas coke, and is used as
fuel. It is greyish-black, porous, and brittle, and contains all the
ash of the coal, most of the sulphur, and small quantities of nitrogen,
hydrogen, and oxygen. The average percentage of carbon is 81,
hence coke is usually classed as a variety of amorphous carbon.

C H O N S Ash H2O

Coal: 5844 3-75 5-99 1-08 1-92 10-05 18-77
Coke: 75-1 049 2-39 0-58 2-63 19-77
The yields from 1 ton of Newcastle coal, in gas -making, are :
12,500 cu. ft. of gas ; HOlb. of tar, yielding 77 Ib. of pitch ; 71b. of
ammonia, and 65-70 per cent, of the weight of the coal is left as coke.
A hard variety of coke for metallurgical purposes (e.g., blast
furnaces) is prepared by carbonising coal in coke-ovens. The old
*' beehive " oven consists of a covered mound of brickwork, in which

Fia. 338.— Coke-ovens.

the coal is partly burnt in a limited supply of air, as in charcoal
burning. The high temperature produced carbonises the rest of the
coal, and all the volatile products are lost. In modern " recovery
ovens," e.g., the Otto or Simon -Carves ovens, the coal is heated in
closed fireclay retorts, O, 24 ft. long and 2 ft. wide, by flues, T,
passing between them in which part of the gas evolved, mixed with
preheated air, is burnt. The gas from the ovens (Fig. 338) is cooled
to separate tar, scrubbed to remove ammonia, and part is burnt in
the flues. The coke is pushed out by rams, and quenched. On
account of the value of the tar, benzene, and ammonia, the use of
recovery ovens is rapidly replacing the wasteful " beehives."

If powdered coke is mixed with pitch or tar-oil, moulded, and
strongly heated in closed retorts, a compact variety of amorphous
carbon, which is a good conductor of electricity, is obtained. This
process is used in the manufacture of carbon electrodes for electric
furnaces.

xxxin CARBON AND THE HYDROCARBONS 685

EXERCISES ON CHAPTER XXXIII

1. Give a concise account of the properties of the two crystalline forms
of carbon. How may graphite be obtained from amorphous carbon ?

2. How would you determine whether a given specimen of carbon
was (a) graphite, (b) diamond, (c) amorphous carbon ? How is graphite
purified, and for what purposes is it used ?

3. What method was used to convert charcoal into diamonds ?
What is supposed to be the condition under which transformation
occurs ?

4. What are the modifications of amorphous carbon ? How are they
made, and for what purposes are they used ?

5. Describe experiments illustrating the property of charcoal of
absorbing gases and dissolved substances. What general connection
is there between the properties of a gas and the extent to which it is
absorbed by charcoal ?

6. How has coal been produced ? What common varieties of coal
are recognised, and what differences are there in their chemical com-
positions ?

7. How are carbides prepared, and what is the action of water on
these substances ?

8. What are saturated and unsaturated hydrocarbons ? Describe
the preparation and properties of one typical member of each of these
groups.

9. What reactions are supposed to take place in the combustions of
hydrocarbons ? How may a mixture of hydrogen and methane be
analysed by the method of fractional combustion ?

10. How are (a) ethylene, (b) acetylene, prepared ? What is the
action of concentrated sulphuric acid on these substances ? How may
they be converted into alcohol ?

11. What is meant by substitution ? How are the facts of sub-
stitution at variance with the electrochemical theory of Berzelius ?

12. How is coal gas manufactured and purified ? What is the com-
position of genuine coal gas ?

13. What is coke ? How is it made on the large scale, and for what
purposes is it used ?

14. What is the calorific power of a fuel ?

15. Forty c.c. of a mixture of hydrogen, methane, and nitrogen were
exploded with 10 c.c. of oxygen. After cooling, the residual gas
measured 36-5 c.c. On treatment with caustic potash the volume
diminished to 33-5 c.c., and on treatment with alkaline pyrogallol to
32 c.c. Calculate the percentage composition of the original mixture.

16. Calculate from the following data the composition of a mixture
of methane, ethane, and hydrogen : vol. of gas taken, 53-5 c.c. ; vol.
of oxygen added, 250 c.c. ; vol. after explosion, 194-95 c.c. ; residue
after treatment with potash, 138-35 c.c.

CHAPTER XXXIV .

OXYGEN COMPOUNDS OF CARBON, ETC.

The oxides of carbon. — Three oxides of carbon, all gaseous at the
ordinary temperature, are definitely known :

Carbon dioxide, or carbonic anhydride, C02, the anhydride of the

hypothetical carbonic acid, H2CO3.
Carbon monoxide, or carbonic oxide, CO, the anhydride of formic

acid, H2CO2.
Carbon suboxide, C302, the anhydride of malonic acid, C3H404.

The oxides C4O3, C8O3, and C12O9 have been described, but their
existence is doubtful. The monoxide and dioxide are the only oxides
of carbon of practical importance.

CARBON DIOXIDE, C02.

Carbon dioxide, C02. — This gas, first prepared by Van Helmont
in 1630 (p. 30), examined by Joseph Black in 1755 (p. 35), and
more fully by Bergman (1774), was clearly recognised as an oxide
of carbon by Lavoisier in 1785. Lavoisier determined its composi-
tion by burning charcoal and the diamond in oxygen, showed that it
combines with bases to form salts, as had been discovered by Black,
and called it acide carbonique. It was long known as carbonic acid
gas.

Carbon dioxide issues in abundance from the earth in certain
localities, such as the Valley of Death (Java) and the Grotto del
Cane (Naples). It occurs in many mineral waters, such as those of
Selters, Vichy, and the Geyser Spring of Saratoga. By the combus-
tion of coal and other carbonaceous fuels, large quantities of carbon
dioxide pass into the atmosphere. The latter contains about 3 vols.
of CO2 in 10.000. Carbon dioxide is formed during respiration, as may
be shown by blowing expired air through lime-water, which becomes
turbid. The fermentation of sugar, in the preparation of beer and
wine, produces carbon dioxide and alcohol : C6H1206 = 2C2H5-OH
(alcohol) -f- 2CO2. Other kinds of fermentation and the decay of
organic matter also produce carbon dioxide.

EXPT. 274. — Dissolve 10 gm. of glucose in 250 c.c. of warm water,
in a flask fitted with a tube dipping into lime-water (Fig. 339). When

686

CH. xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 687

the temperature falls to 30° add a little yeast, and allow the apparatus
to stand for a day or two. The contents effervesce, and bubbles of
gas pass through the lime-water, rendering it milky. The liquid may
be distilled (p. 93), when weak alcohol passes over.

Large quantities of carbon dioxide produced by fermentation in
breweries are collected and liquefied by compression. The liquid
is sold in large steel cylinders, from which the gas may be taken by
standing the cylinder upright with the valve above. If the cylinder
is laid on its side, and the valve opened, a jet of liquid carbon dioxide
issues from it, which, owing to further cooling by rapid evaporation,
at once freezes to a snow-like solid. The latter may be collected by
firmly tying a canvas bag to the jet, and intermittently opening the
latter fairly widely. The solid may
be handled with a horn spoon ; if
pressed between the fingers it pro-
duces painful blisters.

The boiling point of carbon
dioxide is —56° under 5-3 atm.
pressure. The sublimation point of
the solid at atmospheric pressure
is - 78-2° ; this temperature is
attained in a mixture of solid
carbon dioxide and ether, which is
a convenient cooling agent in the
laboratory, and' may be contained

J a , A £ FIG. 339. — Carbon Dioxide by

in a Dewar flask. A second form of Fermentation.

the solid appears to exist under

high pressure. The liquid cannot exist under atmospheric pres-
sure.

EXPT. 275. — Cut a- circular groove in a piece of board, and fill it
with mercury. Place over the whole a mixture of solid carbon dioxide
and ether, by means of a horn spoon. The mercury rapidly freezes.
Knock out the ring of solid mercury, and suspend it by a glass hook
in a jar of water. A thick ring of ice is formed, and the mercury melts.

If solid carbon dioxide is sealed up in a strong glass tube, it melts
under pressure to a liquid. If the tube is warmed gently, the liquid
expands very rapidly, until at 31° the meniscus disappears. At
the same instant the tube is filled with a flickering fog, which at once
vanishes. On cooling the reverse changes occur : 31° is the critical
temperature of carbon dioxide ; the critical pressure is 72 -85 atm.
(p. 170).

Preparation of carbon dioxide. — Carbon dioxide is prepared in the
laboratory by the action of acids on carbonates. The very unstable
carbonic acid, H2CO3, is probably an intermediate product : 2H* -f-
C03" ^± H2C03 ^± C02 + H20.

688 INORGANIC CHEMISTRY CHAP.

EXPT. 276. — Pieces of marble and dilute hydrochloric acid,
in a Woulfe's bottle or Kipp's apparatus, are generally used
for the preparation of carbon dioxide : CaCO3 + 2HC1 = CaCl2 +
CO2 + H2O. The gas is washed with a little water, or passed through
a solution of sodium bicarbonate, to eliminate acid spray, and is then
collected by downward displacement, since it is 1-53 times as heavy as
air. If required free from air, the gas is collected over mercury.

If dilute sulphuric acid is added to marble, the latter soon becomes
coated with sparingly soluble calcium sulphate, CaSO4,2H2O, and the
action ceases. If finely -powdered chalk is used instead of marble, the
reaction is complete, but frothing occurs. Marble or chalk dissolves
readily in concentrated sulphuric acid if a little water is added, since
the calcium sulphate forms a soluble acid sulphate, CaH2(SO4)2. To
remove sulphur dioxide, which is a common impurity, the gas is passed
through potassium permanganate solution.

Pure carbon dioxide is obtained by heating pure sodium bicarbonate :
2NaHCO3 = Na2CO3 + CO2 + H2O ; by the action of dilute sul-
phuric acid on pure sodium carbonate :

Na2C03 + H2S04 = Na2S04 + CO2 + H2O ;

or by heating a mixture of 1 part of sodium carbonate with 3 parts of
potassium dichr ornate.

Carbon dioxide is evolved on heating all carbonates except the
normal carbonates of the alkali metals and barium carbonate.
E.g., chalk, limestone, marble, magnesia alba, etc., give off carbon
dioxide at a red heat : CaCO3 ^± CaO -f C02. The gas is therefore
produced in lime-burning (p. 841).

An impure gas, mixed with nitrogen, is formed by passing a slight
excess of air over red-hot coke or charcoal : C + O2 = C02. If
this gas is passed into a concentrated solution of potassium carbonate,
the carbon dioxide is absorbed, with production of bicarbonate. On
heating the solution, the carbon dioxide is expelled, free from nitro-
gen, leaving a solution of potassium carbonate, which is used over
again : K2CO3 + C02 + H2O ;=± 2KHC03.

Baking-powder contains sodium bicarbonate and tartaric acid, which
do not react when dry. In presence of water, carbon dioxide is evolved,
the bubbles of which are expanded by heat on baking :

2NaHCO3 + C4H6O6 (tartaric acid) = Na2C4H4O6 + 2CO2 -f 2H2O.
Health salt is a similar mixture. The fermentation produced by yeast
in the baking of bread forms carbon dioxide, which gives the dough
a spongy texture.

Properties of carbon dioxide. — Carbon dioxide is a colourless gas
with a faint pungent smell and an acid taste. It extinguishes a
burning taper, sulphur, phosphorus, etc. ; air containing 2 J per cent,
by volume of carbon dioxide will not support the combustion of a

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 689

taper, although 18 J per cent, of oxygen is still present. The gas is
therefore used in extinguishing fires.

EXPT. 277. — Ignite a little benzene in a porcelain dish, and decant
over it a large bell-jar of carbon dioxide. The flame is extinguished.

Fire extinguishers consist of a strong metal vessel containing a
solution of sodium carbonate, with a glass tube of sulphuric acid inside.
By means of a rod attached to a knob outside, the glass tube may be
broken, and the mixture of liquid and gas then issues forcibly from the
nozzle.

Carbon dioxide does not support respiration ; animals die in it
from suffocation, but the gas is not poisonous, and if oxygen is taken
in time recovery with no ill-effects follows.

Burning sodium, potassium, and magnesium continue to burn
in carbon dioxide, with separation of pure carbon : CO2 + 2Mg ==
2MgO + C.

EXPT. 278. — Burn a piece of magnesium ribbon in a jar of dry carbon
dioxide. Treat the residue with dilute sulphuric acid ; magnesia
dissolves, and black specks of carbon are seen floating in the liquid.

A mixture of solid carbon dioxide and magnesium powder burns with
a brilliant flash when ignited, leaving magnesia and carbon.

A characteristic reaction of carbon dioxide is the formation of a
white precipitate of calcium or barium carbonate when the gas is
passed through, or shaken with, lime- or baryta-water. The calcium
carbonate dissolves in excess of carbon dioxide, but barium carbonate
is insoluble (p. 206). Sulphur dioxide also gives a white precipitate
(calcium sulphite, CaS03) with lime-water, but is absorbed by
potassium permanganate solution.

Carbon dioxide is fairly soluble in water (p. 97) ; the latter, at
15°, dissolves about its own volume of the gas. Under pressures
greater than 4—5 atm., at the ordinary temperature, the solubility
increases at a slower rate than the pressure (i.e., according to Henry's
law). On lowering the pressure, the gas escapes with vigorous
effervescence, although the liquid remains supersaturated, and
evolves gas slowly for some time. If the liquid is stirred, or if
porous solids such as sugar or bread-crumbs are thrown into it,
brisk effervescence results. The whole of the carbon dioxide dissolved
in water is expelled on boiling.

Aerated waters (e.g., soda-water) are charged with carbon dioxide
under pressure ; " sparklets " are small iron bulbs containing liquid
carbon dioxide.

Carbonic acid. — The aqueous solution of carbon dioxide has a
faintly acid taste, and turns litmus a port wine red colour. If the
amount of dissolved gas is increased by pressure, the litmus turns

Y Y

690 INORGANIC CHEMISTRY CHAP.

bright red. On boiling, carbon dioxide escapes, and the blue colour
is restored.

A portion of dissolved gas appears to be combined with water to
form carbonic acid, H2C03, and the solution shows very feebly acidic
properties. It appears to be only about one-fifth the strength of
acetic acid ; the latter displaces carbon dioxide from carbonates.
Carbonic acid obeys Ostwald's dilution law, and the dissociation
constants have been given as :

[IT] x [HC03"]/[H2C03] = 3-04 x 10~7 at 18° ;
[IT] x [CO3"]/[HCO3'] =6 x 10"11 at 25°.

From theoretical considerations one would expect carbonic acid
to be stronger than formic acid, H-CO-OH, since the addition of a
hydroxyl group, forming HO -CO -OH, should increase the acidic
properties. It is found that the neutralisation of carbonic acid by
alkali, with phenolphthalein as indicator, is not instantaneous, as is
the case in ionic reactions, so that it is assumed that less than 1 per
cent, of the carbon dioxide is hydrated. The hydration reaction :
CO2 -f- H2O :=^ H2C03, requires time. If the hydrogen ions in the
solution are referred, not to the total CO2, as above, but to the
hydrated part, H2C03, carbonic acid is found to be twice as strong
as formic acid.

Carbon dioxide is more soluble in alcohol than in water. Since
it dissociates in two stages, it is a dibasic acid and forms two series of
salts :

1. Acid carbonates, e.g., NaHC03, Ca(HC03)2 ;

2. Normal carbonates, e.g., Na2CO3, CaC03.

The structural formula of the acid is written HO -CO -OH ; esters
of the hypothetical orthocarbonic acid, C(OH)4, e.g., ethyl orthocar-
bonate, C(OC2H5)4, are known. The acid H2CO3 is metacarbonic
acid. A crystalline hydrate, CO2,6H2O, is obtained under pressure
at low temperatures.

The normal carbonates of alkali metals are hydrolysed in solution,
and exhibit an alkaline reaction : Na2C03 -f- H20 ±^ NaOH +
NaHCO3. A decinormal solution of sodium carbonate is 3-17 per
cent, hydrolysed at 25°.

Dissociation of carbon dioxide. — At high temperatures, carbon
dioxide is slightly dissociated into carbon monoxide and oxygen :
2C02 ^ 2CO -f- 02 : the number of molecules dissociated per 100
molecules of C02 at different temperatures at atmospheric pressure
is shown below (cf. dissociation of steam, p. 212) :

Temperature : 1027° 1170° 1227° 1292° 2367° 2672° 2743°
Percentage
dissociation : 0-004 0-025 0-04 0-06 21-0 65 76

Deville (1865) found that if a rapid stream of carbon dioxide was

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 691

passed through a porcelain tube heated to about 1300°, and the
issuing gas collected over potash, a small amount of a mixture of
carbon monoxide and oxygen was obtained, indicating a dissociation
of about 0 -2 per cent. The gas is also decomposed by electric sparks,
or the silent discharge ; at 3-5 mm. pressure 65-70 per cent, is
decomposed by the silent discharge.

The composition of carbon dioxide. — The composition of carbon
dioxide may be found directly both by weight and by volume. The
composition by weight is determined by burning a weighed amount of
pure carbon in oxygen, and weighing the carbon dioxide, usually
after absorption.

EXPT. 279. — Weigh about 1 gm. of purified sugar -charcoal into a
porcelain boat. Place the boat, X, inside a hard glass tube, Y, one
half of which is packed with recently-ignited granular copper oxide,
Z (Fig. 340). By means of rubber stoppers fit the tube to the purifying
apparatus consisting of U -tubes A and B, containing broken sticks of
caustic potash, and the absorption apparatus, consisting of the weighed
potash-bulbs, C, containing a concentrated solution of caustic potash,

A B

FIG. 340.— Gravimetric Composition of Carbon Dioxide.

with a calcium chloride tube, D, attached. During weighing, these
are closed by bits of glass rod and rubber tubing. Lay the
tube in an iron tray, lined inside with asbestos, in a "combustion
furnace. Sheets of asbestos are placed over the ends of the tube, to
protect the rubber stoppers from heat radiated from the furnace. The
burners underneath the copper oxide are lighted, and the latter is
heated to redness, a slow stream of oxygen from a gas-holder being passed
through the apparatus. The burners under the boat are now lighted,
and the combustion of the carbon is carried out. The layer of hot
copper oxide oxidises any carbon monoxide which may be formed
to carbon dioxide. Allow the oxygen to pass for a few minutes after
the combustion is finished, to sweep out all the carbon dioxide, then
pass air through to displace the oxygen. Detach the potash-bulbs,
closing them with the pieces of glass rod and rubber tubing as in the
previous weighing, cool and reweigh. The increase in weight repre-
sents the carbon dioxide formed. Let x = wt. of carbon, y = wt.
of carbon dioxide ; then y — x = wt. of oxygen. .*. carbon/oxygen
in carbon dioxide = x/(y — x).

Dumas andStas (1841) carried out in this way five combustions of

Y Y 2

692 INORGANIC CHEMISTRY CHAP.

natural graphite, four of artificial graphite, and five of diamond.
The results were in agreement, the mean values being as follows :

800 parts of oxygen combine with :

299-94 parts of natural graphite,
299-95 parts of artificial graphite,
300-02 parts of diamond.

Due allowance was made for ash remaining in the boat after the
combustion. The mean value of the equivalent of carbon (0 = 8)
was taken as 2 -9994 ; this was corrected by Scott for the expansion of
the potash solution after it has absorbed carbon dioxide, which makes
a slight difference to the buoyancy correction in the weighings, and
reduces the equivalent to 2 -9984. Roscoe (1882), by the combustion
of Cape diamonds, found 3-0007 (O = 8), which Scott corrects to
2-9993. Richards and Hoover (1915) determined the ratio Na2C03 :
2Ag: : 29 -43501 : 59 -91676. If the values Ag - 107-88, Na =
22-966 (O = 16) are assumed, the equivalent of carbon is then found
to be 3-001. On the basis H = 1, this gives 2-977.

The volumetric composition of carbon dioxide is found, approxi-
mately, in the same apparatus as was used in the case of sulphur
dioxide (p. 491). A piece of pure charcoal is burnt in a confined
volume of dry oxygen, over mercury. After cooling, it is found that
the volume of the gas is practically unchanged. Thus, the number of
molecules of carbon dioxide produced is equal to the number of
molecules of oxygen disappearing, or one molecule of carbon dioxide
contains one molecule of oxygen. The density of carbon dioxide,
relative to hydrogen, is 21-97, hence its molecular weight is 43-94.
This contains, however, a molecular weight of oxygen, viz., 31-76,
so that the difference, 12-18, represents the carbon. Now it is found
that a molecular weight of any volatile carbon compound never con-
tains a smaller amount of carbon than 12 parts, so that 12-18 should
be the atomic weight of carbon, and the formula of carbon dioxide
is CO2. The corresponding value determined by the gravimetric
method is 2-98x4 = 11-92. The difference is appreciable.
Berzelius (1811), who based his value for the atomic weight of carbon
on the volumetric method just described, was therefore in error by
as much as 2 per cent. This result, when pointed out by Dumas,
shook the confidence of chemists in the atomic weights of Berzelius,
but an active revision of these showed that, except in one or two
cases, they were of a high order of accuracy.

The difference arises from the fact that carbon dioxide is more
compressible than oxygen, so that there is a slight contraction when
carbon is burnt in oxygen. Correct values would be found by the
limiting density method (p. 147), but since the complete compressi-
bility curve of carbon dioxide is not known at very low pressures, the
method has been applied to carbon monoxide and to methane,

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 693

which are more nearly perfect gases than the former. The result
with both is C = 11-910 (H = 1), in complete agreement with the
result of the gravimetric method.

Per carbonates. — If a saturated solution of potassium carbonate is
electrolysed at —10° to —15°, with a platinum anode enclosed in a
porous cell, a bluish-white amorphous precipitate of potassium
percarbonate, K2C206, is deposited at the anode. This may be
washed rapidly with cold water, alcohol, and ether, and dri£d over
P2O5. The formation of the salt is represented as follows :

KO-CO-OK -0-CO-OK' O-CO-OK

= 2K'+ =2K+ j

KO-CO-OK -O-CO-OK' O-CO-OK

It is fairly stable at the ordinary temperature when dry, but is
decomposed by water with evolution of oxygen. The sodium salt
cannot be prepared by electrolysis, since sodium carbonate does not
form a sufficiently concentrated solution. By the action of hydro-
gen peroxide on sodium carbonate a crystalline salt is obtained, which
was formerly considered to have the composition Na2CO4 -f-
JH2O2 + H20. It is now regarded as a carbonate containing hydro-
gen peroxide of crystallisation : Na2CO3 + 1 JH2O2.

Potassium percarbonate, prepared by electrolysis, liberates iodine
immediately from a cold solution of potassium iodide, a reaction
considered to be characteristic of a true percarbonate : K2C206 -f-
2KI = 2K2C03 -f- I2. The sodium compound and hydrogen per-
oxide behave alike in liberating iodine only slowly. By the action of
carbon dioxide on a mixture of sodium peroxide and alcohol, sodium
percarbonate, Na2C206, is formed, which combines with sodium per-
oxide to form sodium permonocarbonate, Na2CO4. Both these salts,
however, liberate less iodine than the equivalent of the active oxygen.
A second potassium percarbonate, K2C2O6, is prepared by the action
of carbon dioxide on alcohol and potassium peroxide ; this resembles
the sodium compound, and differs from potassium percarbonate
obtained by electrolysis, in its action on potassium iodide.

Two isomeric percarbonates, therefore, appear to exist :
(a) KO-CO-0-O-CO-OK (electrolytic}', (ft) KO-0-CO-O-CO-OK
(from peroxide).

The compound Na2C04 is represented as NaO-0-CO-ONa. The
salts K2C206 and Na2C04 are derived from perdicarbonic, or
percarbonic, acid, analogous to perdisulphuric acid (p. 520), and
permonocarbonic acid, corresponding with Caro's acid, respectively :
O-COOH O-COOH

0-COOH OH

peicarbonic acid permonocarbonic acid

694 INORGANIC CHEMISTRY CHAP.

O-SO2-OH OSO2-OH

OS02-OH OH

persulphuric acid permonosulphuric acid (Caro's acid)

By the action of phosphoric acid on potassium percarbonate in ether,
an unstable solution of per carbonic acid, H2C2O6, is formed. H2CO4
is not known.

The carbon dioxide cycle. — In very remote geological periods the
atmosphere of the earth was probably very rich in carbon dioxide,
whilst the primary rocks, such as felspar, K20, Al2O3,6Si06, consisted
almost entirely of bases in combination with silica. At high tem-
peratures, silica displaces carbon dioxide from carbonates, forming
silicates. As the temperature fell, the carbon dioxide and water in
the atmosphere began to decompose the silicates, with the formation
of free silica (quartz), aluminium silicates (clay), soluble alkali
carbonates, and bicarbonates of alkaline earths (e.g., potassium
carbonate, and calcium bicarbonate) : K20,Al2O3,6SiO2 -f- CO2 -f-
2H2O = K2CO3 + A12O3, 2SiO2,2H2O + 4SiO2. These soluble
carbonates (e.g., K2CO3) were partly retained in the soil formed
by this weathering, or pneumatolysis, of the primary rocks, and were
partly washed away to the sea.

Meanwhile, the water of the sea had come into equilibrium with
the atmospheric carbon dioxide, and dissolved a portion of it.
The calcium and magnesium bicarbonates were utilised by
marine organisms, which retained the normal carbonates, and set
free half the carbon dioxide, which was again evolved to the atmo-
sphere. When the organisms died, the calcium carbonate of their
shells was deposited in the form of chalk beds, or coral reefs (a process
which is still going on), producing sedimentary rocks. In this way
carbon dioxide was largely removed from the atmosphere, and
stored up in the form of sedimentary rocks. It is estimated that,
at present, about 30,000 times as much carbon dioxide is contained
in rocks as exists free in the atmosphere.

The proportion of carbon dioxide in the atmosphere was thus
considerably reduced, and further diminution occurred as a result
of the growth of green plants under the influence of sunlight. The
partial decomposition of the remains of these early plants led to
the formation of coal deposits, in which the carbon is largely con-
tained in the free state, or as hydrocarbons rich in carbon. The
process of decomposition of carbon dioxide by green plants may
now be considered.

Photosynthesis. — Green plants contain a pigment known as
chlorophyll, which may be extracted by boiling alcohol. This
pigment occurs associated with protoplasm in the form of cor-

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 695

puscles known as chloroplasts, which are the active agents in the
decomposition of atmospheric carbon dioxide by plants under the
influence of sunlight.

In the leaves of green plants are special organs through which
atmospheric water vapour, oxygen, a little nitrogen, and carbon
dioxide in aqueous solution pass into the cell sap. In aquatic
plants the gases are absorbed entirely from solution. Oxygen and
carbon dioxide are also exhaled by plants. Carbon dioxide is
absorbed by all parts of the surface of the plant which contain
chlorophyll, but mainly by the leaves, and it supplies the material
from which the plant builds up its food. It is converted in the
leaves, under the action of light, first into the carbohydrate starch.
The net result of this change may be represented by the equation :

6a€0a + 5*H20 + energy of light = (CQH1005}X (starch) + 6O2.

The production of oxygen from carbon dioxide by the agency of
living green plants under the influence of light was observed by
Priestley, Ingenhouz, and Senebier, at the close
of the eighteenth century ; it is readily demon-
strated by experiment.

EXPT. 280. — Fill a flask with tap water and
insert into the water some sprigs of watercress or
mint. Fit the flask with a cork through which a
funnel passes, fill the latter with water, and invert
in it a test-tube full of water. Expose the flask
to bright sunlight if available, otherwise to bright
daylight. Bubbles of gas are produced on the
leaves, which rise into the test-tube (Fig. 341).
These are readily shown to consist largely of FIG 341 — Produc-
oxygen.

oxide by Green

The mechanism of the reactions by which Plants in Light.

this process is effected in the plant is unknown ;
recent work has cast considerable doubt on all the theories pre-
viously entertained. The chlorophyll appears to absorb the light
energy which is necessary for the reaction, and acts as a photo-
chemical sensitiser. The reaction itself is called a photosynthesis.

The influence of light in promoting chemical changes was met with
also in the union of hydrogen and chlorine (p. 234). In some cases
the invisible ultra-violet rays of the spectrum are most active, and
the violet end of the spectrum (p. 755) often appears to be more chemi-
cally active than the red, or intermediate, portions. Nevertheless,
the name actinic rays, formerly given to the violet and ultra-violet
parts of the spectrum, is inappropriate, since all the rays of the spec-
trum may be chemically active in different reactions.

696 INORGANIC CHEMISTRY CHAP.

The decomposition of carbon dioxide by the chlorophyll granules
of plants is a case in point. It occurs most rapidly in red and yellow
light, which are absorbed by the green chlorophyll. This part of the
solar spectrum corresponds with the position of maximum energy for
high sun, or the wave-length 666/iAi.

Sulphuretted hydrogen is most rapidly decomposed by red light,
and in some cases even the infra-red rays (so-called " heat rays ") are
most active. Light may also retard a chemical reaction : e.g., the
oxidation of alkaline pyrogallol (p. 719) is retarded by violet light,
but accelerated by red light.

The oxygen absorbed by the plant furnishes the energy by which
its ordinary life-processes are carried on, the light energy being con-
cerned only with the photosynthesis. As a result of the vital pro-
cesses, carbon dioxide is exhaled. Growth ceases in absence of
oxygen : it is most rapid at temperatures of 22° to 37°, and ceases
below 0°, or above 50°.

At night, in the absence of light, the photosynthesis is arrested,
and the starch granules in the leaves pass out of the cells through
the sieve vessels into the sap in the form of soluble carbohydrates
such as sugar, C12H220U. The waste water is given off from the
surface of the plant by transpiration.

The growth of plants. — The food of plants is entirely inorganic.
Besides the gases mentioned above, plants require also mineral
matters, which are absorbed in solution from the soil by the roots.
These include combined nitrogen as nitrates, potassium, calcium,
magnesium, and sodium salts, phosphates, chlorides, silica, and
sulphur as sulphates. The normal soil always contains sufficient
amounts of all these, except potassium salts, nitrates, and phos-
phates, which may have to be added in the form of manures, or
fertilisers. Potassium salts are added in the form of nitre (occa-
sionally), potassium chloride or sulphate, or the crude potash
minerals of Stassfurt. Combined nitrogen is supplied in the form
of Chile nitre, ammonium sulphate, blood, guano, and other nitro-
genous animal products, and farmyard manure. It is in all cases
converted before assimilation into nitrates by the activity of micro-
organisms in the soil (cj. p. 563). Phosphates are supplied as soluble
superphosphate of lime, basic slag, bones, or other phosphates
which can be dissolved by the carbonic acid evolved by decaying
vegetable matter (humus) in the soil.

Small quantities of iron, lithium, manganese, etc., also required, are
taken from the soil. Absorption occurs by selective permeation of
the dissolved salts through the membranes of the root-hairs. If plants
are supplied with carbon dioxide, air, and light, and the roots are
immersed in a solution containing the necessary elements (C, H, O,
N(N03'), S(S04"), P(P04'"), Si(Si02 aq.), 01, K, Ca, Mg, Fe) they con-
tinue to grow. No organic matter is required.

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 697

The weights in Ib. of the various mineral substances removed per
acre by different crops are given below (R. Warington, " Chemistry
of the Farm ") :

Ash. N. K2O. CaO. MgO. P2O5. Cl. SiO2. Na2O. S.
( Wheat, 30

\ bushels ..31 33 9-7 1-0 3-7 14-3 0-2 0-5 0-9 2-7
(Straw, 28 cwt. 158 12 18-2 9-2 4-0 8-4 1-7 110-6 2-5 5-1

! Barley, 40
bushels . . 46 35 9-8 1-3 4-0 16-2 0-4 12-0 1-0 2-9
Straw, 22cwt. 100 12 21-6 8-5 2-5 4-4 3-2 51-5 4-2 3-2
SOats, 45
bushels ..54 38 8-5 2-0 3-9 11-8 — 24-8 1-4 3-2
Straw, 26 cwt. 140 14 29-6 9-8 5-3 7-1 5-5 69-3 5-9 4-8
Meadow hay,

l^tons .. 208 49 56-3 28-1 10-1 12-7 16-2 57-5 11-9 5-7
Red clover hay,

2 tons . . 255 102 87-4 86-1 30-9 25-1 9-4 6-8 4-1 9-4
Turnips and
leaves, 17

tons .. 364 120 148-8 74-0 9-5 33-1 22-1 7-7 24-5 20-9

Mangels and
leaves, 22
tons .. 690 147 262-5 53-3 46-9 49-1 90-4 25-0 140-6 14-0

By the activity of green plants, and marine organisms, therefore,
the carbon dioxide content of the atmosphere tends to be reduced.
We must now consider those processes which tend to increase the
atmospheric carbon dioxide. These are combustion and respiration.

Respiration. — Early experimenters, such as Mayow, Scbeele,
Priestley, and Lavoisier, were all aware of the great similarity
between combustion and respiration. Lavoisier pointed out that
the oxygen breathed into the lungs oxidises the carbonaceous
materials of the blood, producing carbon dioxide, which is exhaled,
and that animal heat is the result of this chemical process of oxida-
tion.

The oxygen passes into the lungs ; these consist of hollow sacs,
the surfaces of which are separated from the blood-vessels by thin
walls, through which the interchange of dissolved oxygen and
carbon dioxide occurs. A certain amount of respiration takes place
through the skin : this process is small in man, but is marked
in animals such as frogs. In the case of fish, dissolved oxygen
is absorbed by the gills.

The blood contains red corpuscles, composed of protoplasm with
a colouring matter known as haemoglobin ; the latter contains iron
in the form of organic compounds, but its exact composition is yet
unknown. Haemoglobin absorbs oxygen, producing a bright red
substance, which exists in the blood of the arteries, passing from
the lungs to the tissues. In the latter, the loosely-combined
oxygen is absorbed, and oxidation processes occur. These are the
source of animal heat and energy, and one of the products is carbon

698 INORGANIC CHEMISTRY CHAP.

dioxide, which remains in solution as carbonic acid or bicarbonates.
The de-oxygenated blood corpuscles have now a dark purple colour,
and part of the blood containing them passes back to the heart
by the veins, to be pumped to the lungs for re-aeration.

The volume of air passing into the human lungs at each inspiration?
or the tidal air, amounts to about 500 c.c ; in forced respiration it may
reach 1640 c.c. The stagnant air, which remains in the lungs, and
mixes with the tidal air, is about 1640 c.c. The expired air contains
by volume 5 per cent, more carbon dioxide and 5 per cent, less oxygen
than the inspired air. It amounts to 400 cu. ft , or 11,200 litres, per
twenty -four hours, and conveys away from the organism about 9 ounces
of water, and 8 ounces of carbon as carbon dioxide, Normal respira-
tion in man occurs eighteen times per minute.

The expansion and contraction of the lungs, by which respiration
occurs, are brought about by movements of the ribs, the muscles of which
are controlled by a nervous centre situated in the medulla oblongata,
or lower portion of the brains This nervous centre is stimulated by the
carbonic acid dissolved in the arterial blood passing through it, and
the activity of the carbon dioxide appears to be due solely to its acidity,
or the concentration of hydrogen ions in the blood. To maintain this
acidity constant within very narrow limits is the function especially
of the kidneys ; the carbonic acid is expelled in the lungs in the form
of carbon dioxide.

In consequence of the activities of plants and animals, the first
absorbing carbon dioxide from the atmosphere, retaining the carbon
and excreting the oxygen, and the latter absorbing oxygen and
excreting carbon dioxide, a kind of balance is maintained between
the proportions of oxygen and carbon dioxide in atmospheric air.

Atmospheric carbon dioxide.— Normal outdoor air contains about
3 volumes of carbon dioxide per 10,000. The average figures for
air at Kew are 243 (minimum)-3-60 (maximum). On Mont
Blanc the figures are 2' 62 at an altitude of 1080 m., and 2-69 at an
altitude of 3050 m. In crowded towns, and especially in rooms
not sufficiently ventilated, the proportion of carbon dioxide may
rise to 0-3 per cent, by volume. The continued breathing of air
containing 0-2 per cent, of CO2 is injurious (Angus Smith). The
" stuffiness " of badly ventilated spaces is chiefly the effect of the
water vapour exhaled by the lungs, which tends to saturate the
stagnant air, and impedes the evaporation of perspiration.

The total amount of carbon dioxide in the atmosphere corre-
sponds with about 600,000 million tons of carbon. The sources of
atmospheric carbon dioxide are : respiration of animals and plants,
combustion, fermentation, putrefaction, the soil (worms, decay,
and gas of volcanic origin), mineral springs, volcanic activity, and
lime-burning. Atmospheric carbon dioxide is diminished by : absorption

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 699

by the sea, photosynthesis by green plants, and the weathering
of siliceous rocks (1-62 X 109 tons of CO2 per annum). On the whole,
the proportion of carbon dioxide in the atmosphere appears to be
slowly increasing, and slight changes of climate may be due partly
to this cause.

In the estimation of atmospheric carbon dioxide, a measured
volume of air may be drawn by an aspirator, first through a drying
tube containing pumice soaked in sulphuric acid, and then through
a weighed tube containing soda-lime. This is followed by a tube of
pumice and sulphuric acid to absorb moisture given off in the soda-
lime tube, and the last two tubes are weighed together. A more
convenient process is Pettenkofer's method. A measured volume of
standard baryta water is shaken with a known volume of the air
in a large (8-10 lit.) bottle, and the excess of baryta titrated
with standard acid and phenolphthalein : Ba(OH)2 -(- CO2 =
BaCOg + H2O. Absorption is more rapid with a hot solution of
baryta.

EXAMPLE. — Volume of air taken in battle = 2360 c.c. at 15° and
762 mm. 20 c.c. of baryta water required 18 c.c. JV/20HC1 (1 c.c. =
0-558 c.c. CO2) beforehand 15-8 c.c. after, shaking with the air. Thus,
volume of CO2 at S.T.P. in the sample of air = (18-0 — 15-8) X
0-558 = 1-228 c.c. Volume of sample at S.T.P.

- 236° >< m X-SI = 2237 c-c- ;

1-228 x 100
hence percentage of CO2 by volume = - oo^o - = 0*055.

CARBON MONOXIDE, GO.

Carbon monoxide, CO. — Lassone (1776) obtained an inflammable
gas by heating charcoal with zinc oxide ; Priestley (1796) substituted
iron-scales (Fe3O 4) for zinc oxide. The latter experimenter considered
the gas to be phlogisticated water, the water supposed to exist in
the calx having combined with the phlogiston of the charcoal. These
experiments were quoted as evidence against Lavoisier's anti-
phlogistic theory, according to which carbonic acid should have
been formed. Cruickshank in 1800 found, however, that the gas
was not inflammable air (hydrogen), but an oxide of carbon con-
taining less oxygen than carbonic acid, and Clement and Desormes
showed that it could be formed by passing the latter over heated
charcoal. Dalton (1808) found that the gas requires half its volume
of oxygen for combustion, and then forms carbonic acid : its for-
mula is therefore CO.

Carbon monoxide occurs in coal gas and in some volcanic gases.
It is formed during the combustion of charcoal or coke in a limited

700 INORGANIC CHEMISTRY CHAP.

supply of air ; the blue flames seen on the top of a clear fire consist
of burning carbon monoxide. The presence of carbon monoxide
in furnace gases is evidence of improper air supply, and its estima-
tion in flue gases therefore affords a useful check on the furnace
efficiency. Poisoning by the fumes of burning charcoal, described
by Hoffmann in 1716, is due to . carbon monoxide, which- is a
dangerous poison.

The production of carbon monoxide in a fire was formerly sup-
posed to be due to the reduction of the carbon dioxide, formed from
the lower portions of the glowing fuel and the entering air, by
passing through the incandescent mass of carbon : C -f- 02 =
CO2 ; C02 + C = 2CO. The monoxide burns on the top of the
fire, where an excess of air is present. The researches of Dixon and
H. B. Baker, however, point to carbon monoxide as a primary
product in the combustion of carbon : .20 -f- 02 = 2CO. If
carbon, carefully dried, is heated in oxygen dried by prolonged
exposure to phosphorus pentoxide, principally carbon monoxide
is formed according to Baker. Wheeler, however, states that both
carbon monoxide and carbon dioxide are formed simultaneously
under these conditions.

The reduction of carbon dioxide by carbon proceeds somewhat
slowly, so that equilibrium : C + C02 ;=± 2CO, is not usually attained
in the combustion of carbon, and the composition of the resulting
gas is variable. The following table contains the equilibrium values
at atmospheric pressure for various temperatures.

C02 + C z± 2CO

Per cent. Per cent.

Temperature. CO2 by vol. CO by vol.

850° 6-23 93-77

900° 2-22 97-78

950° 1-32 98-68

1000° 0-59 99-41

1050° 0-37 99-63.

1100° 0-15 99-85

1200° 0-06 99-94

The formation of a flame of burning carbon monoxide when a
diamond burns in a blast of air was noticed by Macquer in 1771 ;
large quantities of carbon monoxide are also formed when a blast
of air is forced through a thin bed of incandescent coke. The reduc-
tion of carbon dioxide by carbon occurs with appreciable velocity
only at temperatures higher than 600°. The amount of monoxide
formed in equilibrium increases with the temperature. The reverse
reaction : 2CO = C02 + C, was demonstrated by Deville (1864),
who observed the deposition of carbon on a narrow, silvered copper

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 701

tube placed axially in a strongly-heated porcelain tube through which
carbon dioxide was passed. The copper tube was cooled by a
stream of water.

EXPT. 281. — Pass a slow current of carbon dioxide over pieces of
charcoal heated to redness in an iron tube (Fig. 342). The carbon
dioxide is removed from the issuing gas by a tube of soda -lime, and
the monoxide may then be burnt at a jet.

Carbon monoxide is formed by heating charcoal with zinc, iron,
or manganese oxides : C 4- ZnO = Zn + CO, or with chalk or

FIG. 342. — Carbon Monoxide from Carbon Dioxide and Carbon.

barium carbonate : BaCO3 -f- C = BaO -f 2CO. It is also pro-
duced by passing carbon dioxide over zinc dust or iron filings heated
to redness in a glass tube : CO2 -f- Zn = ZnO + CO. Calcium,
magnesium, and the alkali-metals, on the other hand, lead to separa-
tion of free carbon : 2Ca + C02 = 2CaO + C, and 4K + 3CO2 =
2K2C03 + C.

Preparation of carbon monoxide. — Although carbon monoxide is
produced on the large scale by passing carbon dioxide over heated
carbon (see p. 1002), in the laboratory it is more conveniently
prepared by heating formic acid (or sodium formate), oxalic acid,
or potassium ferrocyanide, respectively, with concentrated sulphuric
acid.

702 INORGANIC CHEMISTRY CHAP.

The gas obtained from formic acid is almost perfectly pure :
H-COOH = H2O -f- CO ; a trace of sulphur dioxide may be formed
in this, and in the following reactions, bv reduction of the sulphuric
acid : H2S04 + CO = CO2 -f SO2 + H2O, but this is removed by
washing with caustic soda.

EXPT. 282.— Concentrated sulphuric acid is heated to 100° in a flask,
and concentrated formic acid dropped in from a tap-funnel (Fig. 333).
Cold concentrated sulphuric acid may also be dropped on dry sodium
formate in a flask. The gas is washed with caustic soda, dried with
phosphorus pentoxide, and collected over mercury. It is then pure.

Oxalic acid, when gently heated with concentrated sulphuric
acid, evolves a mixture of equal volumes of carbon monoxide and
dioxide : (COOH)2 = CO + CO2 + H20. The carbon dioxide is
easily removed by washing with caustic soda.

EXPT. 283. — Twenty-five gm. of crystallised oxalic acid (C2H2O4,2H2O)
are covered in a flask with concentrated sulphuric acid. On heating
gently, a brisk evolution of gas occurs. . Fill a long tube divided into
two parts by a paper label, and fitted with a stopcock, with the gas.
Then attach a wash-bottle containing caustic soda solution to the
generating apparatus, and collect jars of carbon monoxide over water.
Note : carbon monoxide is very poisonous.

Admit a little caustic soda solution to the long tube of mixed gas,
shake, and then open the stopcock under water. The latter rushes in
and fills half 'the tube. Hence the gas contained half its volume of
carbon dioxide.

Potassium ferrocyanide, on heating with ten times its weight of
concentrated sulphuric acid in a large flask, evolves nearly pure
carbon monoxide, but the reaction is usually somewhat violent :

K4Fe(CN)6 + 6H2S04 + 6H2O = 2K2S04 + FeS04 +
3(NH4)2S04 + 6CO.

The gas evolved in the later stages of the reaction is not pure.

Carbon monoxide is produced by withdrawing the elements of
water from formic acid ; this is effected by concentrated sulphuric
acid, or by the catalytic action of metallic rhodium. The reverse
reaction, i.e., the synthesis of formic acid, is effected bv the action
of the silent discharge: CO + H20 ;=± H-CO-OH, and sodium
formate is produced by passing carbon monoxide over caustic soda,
or soda-lime, at 200° : NaOH -f CO = Na-COOH. Carbon
monoxide is, therefore, the anhydride of formic acid. The anhydride
of oxalic acid, C2O3, does not exist, but breaks up at once into
CO -f C02.

Properties of carbon monoxide. — Carbon monoxide is a colourless
gas with a peculiar faint smell. It is very poisonous, 10 c.c. per kg.

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 703

weight of an animal produces death, and the inhalation of air
containing 1 vol. of CO in 800 vols. is fatal in half an hour. Coal
gas (especially modern gas, which contains water-gas) owes its
poisonous properties to the carbon monoxide it contains. The
fumes of burning charcoal are lethal for the same reason.

The poisonous action of carbon monoxide depends on the absorption
of the gas by the haemoglobin of the blood, forming bright -red carboXy-
haemoglobin, which is a very stable substance. Oxygen is unable to
displace carbon monoxide from the compound, and the animal dies
because of lack of oxygenation of the blood and tissues. The absorption
spectra (p. 762) of oxy-hsemoglobin and car boxy -haemoglobin are
similar but distinct, so that poisoning with carbon monoxide may
readily be detected by examining the absorption spectrum of the blood.
In cases of poisoning, artificial respiration and administration of oxygen
should be resorted to at once, the patient being kept warm and at rest ;
alcohol may be given if there is a tendency to fainting.

Carbon monoxide is liquefied with difficulty ; its critical tem-
perature is — 140°. The liquid boils at — 193°, and solidifies at
- 200°.

The gas is sparingly soluble in water, but is readily absorbed by
a solution of cuprous- chloride in hydrochloric acid, a white
crystalline compound, CuCl,CO,2H2O, being formed. Water or
ammonia must be present ; a solution of cuprous chloride in dry
alcohol does not absorb the gas.

The composition of carbon monoxide is determined by passing
it over heated copper oxide, the carbon dioxide formed being
absorbed in weighed potash-bulbs. The normal density of the gas
is 1/2504, hence the relative density is 13*9, and the molecular
weight 27-8 (approximately). The gas when mixed with half its
volume of oxygen and exploded yields its own volume of carbon
dioxide. The formula is therefore CO.

Many metals form compounds called carbonyls, with carbon
monoxide: Co(CO)3, Co2(CO)8, Ni(CO)4, Fe(CO)4, Fe(CO)5,
Fe2(CO)9, Mo(CO)6, Ru(CO)*. Carbon monoxide penetrates heated
iron and may escape through the iron flues of stoves burning with
an insufficient supply of air. Carbon monoxide also combines
directly with chlorine, forming carbonyl chloride (phosgene), COC12.

Combustion of carbon monoxide. — Carbon monoxide burns in
air or oxygen with a beautiful blue flame, forming carbon dioxide.
The gas is also a powerful reducing agent, and when passed over
heated metallic oxides it abstracts the oxygen contained in them,
leaving the metal : PbO + CO = Pb + CO2. Carbon monoxide
is the active agent in a number of metallurgical processes
(cf. the blast furnace). It reduces iodine pentoxide at 90°, with
liberation of iodine : I205 + 5CO = I2 -f 5C02. This reaction

704 INORGANIC CHEMISTRY CHAP.

may be used for the estimation of carbon monoxide in gases. A
mixture of carbon monoxide and hydrogen may be analysed in
this way, or by passing the mixture with oxygen over palladium-
asbestos ; only the h}Tdrogen is oxidised. If a mixture of carbon
monoxide and methane is passed over copper oxide at 250°, only
the carbon monoxide is oxidised. If gas containing only O05 per
cent, of CO is shaken with a solution of palladious chloride, a black
precipitate of palladium is produced.

The explosion of carbon monoxide with oxygen. — A mixture of two
volumes of carbon monoxide and one volume of oxygen explodes when
lighted in the ordinary way. H. B. Dixon in 1880 found, however,
that if the gases are carefully dried by exposure to phosphorus
pentoxide, they cannot be exploded in a eudiometer, although
combination occurs locally in the path of the electric sparks. If a
trace of moisture, or of any gas which contains hydrogen, and so
produces water on combustion in oxygen (CH4, H2S, etc.), is added,
the mixture can be exploded by a spark. M. Traube (1885) found
that a burning jet of carbon monoxide, dried by passing through
towers containing glass beads wetted with very concentrated
sulphuric acid, is extinguished when plunged into a jar of oxygen
containing very strong sulphuric acid which has been dried by stand-
ing for a few hours carefully stoppered.

Girvan (1903) finds that 1 molecule of water in 24,000 of the gas is still
active. The . maximum effect is produced by 4-5 per cent, of water
vapour.

The catalytic influence of moisture in this (and other similar) reactions
is still somewhat obscure. Since carbon monoxide readily reduces
steam at high temperatures : CO + H2O ^ CO2 + H2, Dixon supposes
that this reaction first occurs, and that the hydrogen then combines with
the oxygen present to reproduce water : 2H2 + O2 = 2H2O, and so on.

Catalytic effect of moisture. — Numerous cases of the cata-
lytic effect of moisture are known. Dry chlorine does not combine
with dry metals, except mercury. Dry carbon monoxide and
oxygen do not explode on sparking. In the absence of moisture,
to the extent produced by prolonged drying over phosphorus
pentoxide, carbon combines only slowly with oxygen on heating ;
ammonium chloride and calomel volatilise on heating without
dissociation ; ammonia and hydrogen chloride do not combine
on mixing ; and sulphur and phosphorus may be distilled unchanged
in oxygen. Nitrogen trioxide, after prolonged drying in the liquid
state over P2O5, volatilises as N406 ; in presence of a minute trace
of moisture this instantly dissociates into NO and NO2. The
boiling point of liquid N466 is also raised from — 2° to +43° by
drying for three years. Calomel dried for six months over P205 at

OXYGEN COMPOUNDS OF CARBON, ETC.

705

Hopper

XXXIV

115° will not vaporise at all at 352°, when its usual vapour pressure
is 347 mm.

In some cases the presence of pure water is not sufficient to
catalyse a reaction, but a trace of impurity is needed.

H. B. Baker (1902) found that a mixture of very pure hydrogen and
oxygen from the electrolysis of baryta, if sealed up in glass tubes over
purified P2O5, did combine slowly, after prolonged drying, when the tube
was heated with a flame, or if a spiral of silver wire was heated almost to
the melting point in the gas, but no explosion occurred. The water
produced by the combination was, according to Armstrong's theory
(1885), too pure to form an electrically -conducting circuit, which he
considers necessary for chemical change.

Producer gas. — The gaseous mixture obtained by passing air
through a bed of incandescent coke, consisting principally of nitrogen
and carbon monoxide, is
made for heating pur-
poses, and is called pro-
ducer gas (or air-gas). The
producer consists of a
closed fire-grate in which
coke rests on bars ; it is
often sealed below by
water, and the primary
air is either drawn
through the fuel with a
fan, or forced through by
pressure, the ash-pit then
being air-tight (Fig. 343).
If the gas is burnt with-
out cooling, the total
amount of heat evolved
is the same as if the

carbon were burnt directly to carbon dioxide : usually 30 per
cent, of the heat is lost by the producer gas cooling before it
arrives at the place where it is burnt. Gas-firing is preferred for
many purposes on account of the ease with which it is regulated
and its cleanliness. The air admitted for the combustion of the
producer gas is called secondary air.

If coal is used instead of coke, the gas will be mixed with coal gas,
unless the draught through the producer is downwards, when the coal
gas is decomposed by the incandescent fuel. Otherwise the tar must be
separated from the gas (" suction -gas ") ; with down-draught it is
absent.

Water gas. — If steam is blown through incandescent coke, a

z z

Em. 343.— Gas Producer.

706 INORGANIC CHEMISTRY CHAP.

mixture of carbon monoxide, carbon dioxide, and hydrogen is
formed, known as water gas: (1) C -f H20 ^r CO -f H2 ;
(2) CO + 2H20 ^± C02 + 2H2. The proportion of carbon monoxide
increases as the temperature rises, as is seen from the following
table, giving the results of Bunte :

Percentage Composition of gas
of steam. by volume. CO H2 CO

Temp.
675°

\A.\3\s\Jl.ll.-

posed.
8-8

H2
65-2

4-9

CO2

29-8

CO2
0-16

CO CO+CO2
13-3 0-141

758

25-3

65-2

7-8

27-0

0-29

8-4

0-224

840

41-0

61-9

15-1

22-9

0-65

4-1

0-397

955

70-2

53-3

39-3

6-8

5-80

1-35

0-853

1010

94-0

48-8

49-7

1-5

33-10

0-98

0-972

1060

98-0

50-7

48-0

1-3

36'8

1-04

0-975

1125

99-4

50-9

48-5

0-6

80-8

1-05

0-988

Average water gas has the following composition: H2, 49-17;
CO, 43-75 ; C02, 2-71 ; methane, 0-31 ; N2; 4-00. Its calorific power is
about 350 B.Th.U. per cu. ft., but as it requires only 2-5 vols. of air
for combustion, it gives a very hot flame.

The reactions in the water gas producer absorb heat, hence the
hot coke is gradually cooled by the steam blast, and the amount
of carbon dioxide in the gas increases. When the steam blast has
passed for a certain time (eight to twelve minutes), it is shut off,
and an air blast turned on until the fuel is again heated to bright
redness (one and a half to two minutes). The producer gas formed
in the air-blow is utilised in raising steam, although extra fuel
must be used for this purpose. To keep the temperature as
uniform as possible, the steam blast is passed alternately upwards
and downwards through the producer. In recent types, the fuel
bed is thin, and carbon dioxide is largely formed during the air-
blow.

Semi-water gas is prepared by passing a mixture of steam and
air continuously through incandescent coke, the heat evolved by the
combustion of the carbon with the oxygen of the air being sufficient
to maintain the temperature for the water gas reaction to occur
with the steam. About four times as much carbon is burnt by the
air as reacts with the steam. Mond gas is formed with a large
excess of steam which keeps the temperature low (650°), and allows
of the recovery as ammonia of a larger proportion of the nitrogen
of the coal-slack used than if the coal had been heated in retorts.

Carburetted (" enriched ") water gas is formed by mixing water
gas with hydrocarbons, partly unsaturated, which burn with a
luminous flame. Water gas alone (i.e., a mixture of hydrogen,
carbon monoxide, and nitrogen) burns with a blue, non-luminous

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 707

flame, but may be used with Welsbach mantles for illuminating
purposes, since it gives out a considerable amount of heat on com-
bustion. In the manufacture of carburetted water gas, two towers
packed with chequer-brickwork are placed after the producer.
The first, called the carburetter, and the second, called the super-
heater, are first heated to redness by the hot producer gas from the
air-blow passing down the first and up the second. The water gas
from the steam-blow is now passed through the towers. Into the
carburetter a spray of mineral oil is injected. This vaporises, and
the mixture of water gas and oil vapours then passes through the
red-hot bricks in the superheater, where the oil vapour is decom-
posed, or " cracked," with the formation of permanent gases ri'ch
in ethylene. The gas is then scrubbed and collected. Pintsch
gas is formed by spraying oil into hot retorts and passing the gas
through a condenser, scrubber, and lime purifier.

The compositions of two typical specimens of semi-water gas
(producer gas) are given below, together with an analysis of true
water gas :

CO. H2. CH4. C02. 02. N2.

Dowson gas from coal .. 25-07 18-73 0-62 6-57 49-01

Do. from coke. . 2240 7-00 4-90 0-50 65-20

Mond gas from coal .. 13-20 24-80 2-30 12-90 46-80

Water gas .. .. 39-6 51-9 0-8 4-2 2-9

The calorific power of producer and semi-water gas is very low,
being usually about 125 B.Th.U. per cu. ft., as compared with about
600 for good coal gas. The adulteration of modern coal gas by
water gas has considerably reduced its calorific value.

The following thermal constants are useful in fuel calculations :

(1) 1 Ib. of carbon burning to carbon dioxide evolves 14,544 B.Th.U.

(2) 1 Ib. of carbon burning to carbon monoxide evolves 4351 B.Th.U.

(3) 1 Ib. of carbon reacts with steam to produce water gas

(C + H2O = CO + H2) with the absorption of 4298 B.Th.U.

(4) 1 Ib. of hydrogen burns to liquid water with the evolution of

60,626 B.Th.U.

(5) 1 Ib. of carbon monoxide burning to dioxide evolves 4368 B.Th.U.

Hydrogen from water gas. — The manufacture of hydrogen from
water gas is carried out in different ways : the carbon dioxide is
first removed by washing with lime, and the carbon monoxide
then separated by one of the following processes : —

(1) Washing with cuprous chloride solution, or with hot concentrated

caustic soda under pressure : CO + NaOH = H-COONa (sodium
formate) ;

(2) passing over calcium carbide at 300° ; CO gives CaO and CaCO3 ;

nitrogen forms calcium cyanamide, CaCN2 (p. 544) ;

z z 2

708 INORGANIC CHEMISTRY CHAP.

(3) liquefaction of carbon monoxide by compression and cooling ;

the residual gas contains 2 per cent, of CO, removable by pro-
cess (2) ;

(4) passing over lime, alone or mixed with oxide of iron, heated to

400-500° : CaO + CO + H2O = CaCO3 + H2 ;

(5) mixing with steam and passing over finely-divided nickel or

cobalt at 350-400°, or under 4-40 atm. at 300-600° in presence of
nickel, iron, cobalt, or other catalyst. A little oxygen is added
to maintain the temperature of the catalyst : CO + H2O ^^
CO2 -+- H2. If CO2 is added, the formation of CO from deposited
carbon : C -{- H2O ^ CO -f- H2, is prevented by mass-action.

Carbonyl chloride, or phosgene, COC12. — When a mixture of equal
volumes of carbon monoxide and chlorine is exposed to bright
sunlight, or passed over heated animal charcoal, direct combination
occurs, with the formation of carbonyl chloride, or phosgene, COC12
(Greek, phos, light, and gennac, I produce). This compound, dis-
covered by John Davy in 1811, is a colourless gas with a penetrating
and suffocating odour, and is very poisonous. It is readily liquefied
by cooling, forming a colourless, mobile liquid, b.-pt. 8°. The gas
does not fume in moist air, but is readily hydrolysed by water.
The hypothetical carbonic acid, H2C03, may first be produced :

OH ,OH

C0;

iCl H
H

= 0:C< + 2HC1 = C02 + H20 + 2HC1.
OH XOH

Phosgene is, therefore, the chloride of carbonic acid. When the
gas is passed into a solution of ammonia in toluene, urea is formed,
which may be regarded as the diamide of carbonic acid, or dicarb-
amide, CO(NHg)a: COC12 + 4NH3 = CO(NH2)2 + 2NH4C1. Both urea
and ammonium chloride are precipitated, but may be separated
by warming with alcohol, in which urea is soluble. The alcoholic
solution deposits, on evaporation and cooling, crystals of urea.

The monamide of carbonic acid, C0< , is called carbamic

XNH2
X0-NH4
acid. Its ammonium salt, C0< , is contained, together with

ammonium bicarbonate, NH4HC03, in commercial " carbonate of
ammonia " (p. 801).

Carbonyl bromide, COBr2, is slowly formed from carbon monoxide
and bromine vapour.

Carbonyl sulphide, or carbon oxysulphide, COS. — This compound,
discovered by Than in 1867, is formed when carbon monoxide and

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 709

sulphur vapour are passed through a heated tube : CO -f S ^ COS,
or when sulphur dioxide is passed over red-hot charcoal. It is most
conveniently prepared by the action of diluted sulphuric acid
(5 vols. of H2SO4 to 4 vols. of water) on ammonium thiocyanate,
NH4CNS, at 20°. The unstable thiocvanic acid appears first to be
formed, and is hydrolysed by water : HCNS -f H20 = COS + NH3.
The gas so prepared contains hydrocyanic acid, HCN, and carbon
disulphide. The first is removed by passing through very con-
centrated caustic potash solution ; the latter by passing through
concentrated sulphuric acid followed by a mixture of triethyl
phosphine, P(CHg)3, pyridine, and benzene.

Carbonyl sulphide is a colourless, odourless gas, sparingly soluble
in water, but readily soluble in toluene. It liquefies at 0° under
12 atm. pressure, b.-pt. —50-2°; m.-pt. --138-2°. It is very
inflammable, a glowing chip causing its ignition, and burns with a
blue, slightly luminous, flame. When mixed with oxygen, it ex-
plodes feebly with a spark, even after drying with phosphorus
pentoxide, although neither carbon monoxide nor sulphur
burns when perfectly dry : 2COS -f 302 = 2CO2 -f 2S02. A heated
platinum spiral decomposes the gas without change of volume
into sulphur and carbon monoxide : COS = CO + S (solid).

The aqueous solution of carbonyl sulphide is slowly hydrolysed :
COS -f H20 ^± HO-CO-SH =r C02 + H2S. The intermediate sub-
stance, HO-CO-SH, is thiolcarbonic acid (p. 715). The hepatic
waters of Harkany and Parad, in Hungary, appear to contain
carbon oxy sulphide. Carbonyl sulphide is absorbed by dilute
aqueous or alcoholic potash with the formation of a mixture of
sulphide and carbonate : COS + 4KOH = K2C03 + K2S + 2H2O.

Formic acid, H-CO-OH. — The absorption of carbon monoxide by
heated alkalies, with the production of formates, has already been
mentioned. At 120°, under 3 to 4 atm. pressure, the gas is rapidly
and completely absorbed by a concentrated solution of caustic
soda : NaOH + CO = H-COONa. Large amounts of sodium
formate are prepared by this method. From this, anhydrous formic
acid, H-CO-OH, is obtained cheaply and in quantity. Thirty-five
parts of concentrated sulphuric acid are run into 200 parts of con-
centrated formic acid, with shaking. To this mixture 50 parts of
sodium formate and 50 parts of concentrated sulphuric acid are
added alternately and the liquid distilled.

Formic acid is a colourless liquid, sp. gr. 1-226, b.-pt. 100-6°,
m.-pt. 8-43°, with a pungent odour. It acts violently on the skin,
raising blisters. The acid is contained in red ants (Formica rubra),
and was first obtained from them by distillation in steam. It is
also present in nettles, and in nearly all stinging, organisms.

Formates are powerful reducing agents. If mercuric oxide is
dissolved in dilute formic acid, it goes into solution as mercuric

710 INORGANIC CHEMISTRY CHAP.

formate. This is soon reduced to a white precipitate of mercurous
formate, and finally to grey metallic mercury. The formic acid is
oxidised to carbon dioxide. In presence of ruthenium, rhodium,
and iridium, especially if traces of the sulphides are present, formic
acid decomposes into carbon dioxide and hydrogen : H2C02 =
CO 2 + H2. The reverse reaction occurs on electrolytic reduction
with a clean zinc cathode, or by passing hydrogen through a solution
of a bicarbonate containing palladium or platinum.

If a mixture of sodium formate with one -twentieth of its weight
of caustic soda is heated to 250-260°, hydrogen is evolved, and
sodium oxalate remains : 2HC02Na = (C02Na)2 -f H2. From sodium
oxalate free oxalic acid, (CO2H)2,2H2O, is easily obtained. By the
electrolytic reduction of oxalic acid, glyoxylic acid, H-C()-CO2H,
and finally giycollic acid, H2(OH)OC02H, are obtained in large
quantities. All these organic compounds, therefore, may be
obtained directly from carbon monoxide.

Carbon suboxide, C302. — If malonic acid, CH2(COOH)2, or ethyl
malonate, CH2(COO-C2H5)2, is treated with a large excess of phos-
phorus pentoxide at 300° under 17 mm. pressure, carbon suboxide,
C302, is evolved. The reaction with malonic acid is : CH2(COOH)2=
C3O2 + 2H20 ; that with ethyl malonate is : CH2(COO-C2H5)2 =
C302 + 2H20 + 2C2H4.

The gas evolved is liquefied by cooling, and is fractionated ;
the carbon suboxide boils at 6°. It freezes in liquid air
to a white solid, m.-pt. -- 11 1-3°. The gas has a pungent
odour, and is poisonous. It burns in air with a smoky flame,
and explodes with oxygen when ignited : C302 + 2O2 = 3C02.
The liquid slowly polymerises at the ordinary temperature,
forming a red solid insoluble in water, and the gas decomposes
rapidly on heating or in contact with phosphorus pentoxide. Carbon
suboxide dissolves readily in water, forming a solution of malonic
acid, of which it is the second anhydride, i.e., formed by the removal
of two molecules of water from one molecule of the acid. Its
formula is, therefore, O:C:C:C:0. The gas is readily soluble
in benzene and xylene.

Carbon disulphide, CS2. — Sulphur vapour when passed over red-
hot carbon produces carbon disulphide, CS2, a volatile liquid, the
reaction being endothermic and reversible : C -|- 2S ^ CS2 — 254
kgm. cal. Since heat is absorbed in the reaction, the yield is
improved by working at a high temperature, and the compound is
now largely manufactured in the electric furnace. Carbon disul-
phide was discovered by Lampadius in 1796, by heating pyrites
with charcoal.

In the older process a vertical cast-iron or fireclay retort is set
in a furnace and filled with charcoal (Fig. 344). Sulphur is fed in
through a side tube, a, at the base of the retort, being kept fused

XXXIV

OXYGEN COMPOUNDS OF CARBON, ETC.

711

by the waste heat. The sulphur volatilises, and the vapour passes
over the white-hot charcoal, forming carbon disulphide. The

FIG. 344. — Manufacture of Carbon Disulphide.

vapours pass through a small iron cylinder, d, where sulphur is
deposited, and the carbon disulphide is condensed in a very long
worm-tube cooled by water.

In Taylor's electrical process (1899) a tower 40 ft. high and
16 ft. uTdiameter (Fig. 345) is packed with charcoal or coke from
the top. Below this is a furnace
with four carbon electrodes,
between alternate pairs of which
an arc is struck. The sulphur
in the lower part of the furnace
melts and evaporates, the vapour
passing through the heated coke
above the arc, and forming
carbon disulphide. Fresh coke
and sulphur are added every
twelve hours through the hop-
pers shown. The disulphide
is condensed as before, and
purified by redistillation. It is
further purified by agitation
with mercury, and redistilled FlG 345._Tayior's Electric Carbon Disui-

OVer white wax. Phide Furnace.

EXPT. 284. — A combustion tube packed with small pieces of recently
ignited charcoal is fitted in a sloping combustion furnace as shown in

712 INORGANIC CHEMISTRY CHAP.

Fig. 346. The lower end is connected with bulb tubes surrounded by
ice. When the tube is red hot, bits of sulphur are introduced into the
upper end, which is corked. The sulphur vapour passes over the hot
charcoal, and the carbon disulphide formed (containing sulphur in
solution) is collected in the bulbs.

Properties of carbon disulphide. — Carbon disulphide is a colour-
less, mobile, strongly refracting liquid, which boils at 46°, solidifies
at — 116°, and remelts at — 110°. Its density at 0° is 1-2923.
The liquid is slightly soluble in water, the solubility diminishing
with rise of temperature. One hundred c.c. of water dissolve
0-204 gm. of CS2 at 0°, 0-179 at 20°, and 0-014 at 40°. Carbon
disulphide readily volatilises, and its vapour has an exceedingly
unpleasant odour, which is not removed by careful purification
(Dixon). The vapour ignites at a very low temperature : a test-
tube filled with hot oil held over the liquid in a dish sets fire to the

FIG. 346.— Preparation of Carbon Disulphide.

vapour. The vapour forms a violently explosive mixture with
oxvgen, the most violent explosion being obtained with
2CS2 + 502 = 2CO + 4SO2. Sulphur dioxide, sulphur trioxide,
carbon monoxide, and carbon dioxide are formed. No free carbon
is deposited.

Carbon disulphide, being an endothermic compound, is unstable
at the ordinary temperature. If a little mercury fulminate is
exploded in a tube filled with the vapour, decomposition commences,
with separation of sulphur and carbon, but is not propagated
through the vapour.

Carbon disulphide mixes with absolute alcohol, ether, and oils.
It also dissolves sulphur, white phosphorus, indiarubber, camphor,
resins, etc., and is largely used as a solvent.

The vapour is decomposed by heated potassium : CS2 -f- 4K =
2K2S + C. When chlorine is passed into boiling carbon disulphide
containing a little iodine, carbon tetrachloride, CC14, is formed :
CS2 + 3C12 = CC14 (b.-pt. 77°) + S2C12 (b.-pt. 136°).

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 713

Both products of this reaction are useful, and are separated by
fractional distillation. Carbon tetrachloride is used, under the
name of pyrex, as a grease solvent, and for extinguishing fires.

Carbon disulphide vapour acts as a powerful poison when inhaled :
it is used to kill moths in furs, etc., and mice and rats in grain
elevators.

A mixture of carbon disulphide vapour and hydrogen, when
passed over heated platinised pumice, or nickel at 450°, vields
hydrogen sulphide : CS2 + 2H2 = C + 2H2S (p. 682).

This reaction is used in determining the amount of CS2 in coal gas :
the H2S produced is estimated by passing the gas through a solution of
lead nitrate in sugar syrup, and matching the brown tint of the PbS with
standards.

Carbon disulphide reacts with an ethereal solution of triethyl
phosphine, P(C2H5)3, forming a red crystalline compound

CS - P(C2H5)3

P(C2H5)3,CS2, possibly with the constitution \ /

When the vapour of carbon disulphide is passed over red-
hot copper, carbon is deposited and copper sulphide formed :
CS2 -f- 4Cu = C -f- 2Cu2S. It was in this way that the composition
of the substance was first determined by Vauquelin. A mixture
of the vapour with steam, or sulphuretted hydrogen, when passed
over red-hot copper, gives methane :

CS2 + 2H2O + 6Cu = CH4 + 2Cu2S + 2CuO ;
CS2 + 2H2S + 8Cu = CH4 + 4Cu2S.

From methane, organic substances such as alcohol and acetic acid
may be obtained, so that these reactions allow of the synthesis of such
compounds, carbon disulphide being prepared directly from its element.
(Berthelot, 1856.)

Carbon subsulphide, C3S2. — This compound, corresponding with the
suboxide C3O2, was discovered by Lengyel. It is formed by striking an
arc under carbon disulphide, the cathode being of carbon and the anode
of antimony containing 7 per cent, of carbon. The liquid is then
distilled in vacuo, and the vapour condensed at — 40°. A yellowish-
red solid is formed, m.-pt. — 0-5°. It has the composition C3S2, and
the structural formula is probably S:C:C:C:S, similar to that of
C3O2. The vapour has an offensive odour, and produces a copious flow
of tears. A dibromide, C3S2Br2, is formed directly, and has a not
unpleasant aromatic smell.

Carbon monosulphide, (CS)X, is said to be contained in the brown
powder produced when carbon disulphide is exposed to light. Thio-
carbonyl chloride, CSC12, is formed when a mixture of carbon disulphide,

714 INORGANIC CHEMISTRY CHAP.

chlorine, and a trace of iodine is heated in a sealed tube for some time,
or when a mixture of phosphorus pentachloride and carbon disulphide is
heated in a sealed tube at 100°: PC15 + CS2 = PSC13 + CSC12. It is a
liquid, boiling at 149° with slight decomposition, has a very offensive
odour, and is slowly hydrolysed by water. When treated with nickel
carbonyl, solid (CS)X is formed.

Carbon sulphoselenide, CSSe, and sulphotelluride, CSTe, have been
prepared by striking an arc under carbon disulphide between a graphite
cathode and an anode of graphite and selenium, or tellurium, respectively.
They are yellow and red liquids, respectively.

Thiocarbonic acid. — If carbon disulphide is agitated with a con-
centrated solution of caustic soda it slowly dissolves. The solution
contains sodium carbonate, and a new salt, sodium thiocarbonate,
Na2CS3, which may be regarded as the carbonate in which oxygen
is replaced by the analogous element sulphur :

3CS2 + 6NaOH = 2Na2CS3 + Na2C03 + 3H20.

If a solution of sodium sulphide is used instead of caustic soda, the
reaction is more rapid, and sodium thiocarbonate alone is formed :
Na2S -J- CS2 •= Na2CS3. On adding alcohol, the thiocarbonate
separates as a heavy oily liquid of slightly brown colour. On
treating this with cold dilute hydrochloric acid, free thiocarbonic
acid, H2CS3, separates as a yellow oil, decomposed on warming into
sulphuretted hydrogen and carbon disulphide : H2CS3 =
H2S -f- CS2. A deep red solution and yellow crystals of the ammo-
nium salt, (NH4)2CS3, are formed when carbon disulphide and
concentrated ammonia are allowed to stand together for a few
days. The relationship between these and the corresponding com-
pounds containing oxygen is obvious, and was pointed out by their
discoverer, Berzelius (1825) :

Anhydride . . C00 CS2

Acid . . H2C03 (H2O + C02) H2CS3 (H2S + CS2)

Salt . . KaCO8 (K2O -f C02) K2CS3 (K2S + CS2)

Thiocarbonates are used in destroying phylloxera, a fungus
infesting vines. Carbon disulphide is a poison for this fungus,
but it is too volatile to use directly ; if the plants are sprayed with
a solution of sodium thiocarbonate, this is slowly decomposed by
atmospheric carbonic acid, with liberation of carbon disulphide.

If carbon disulphide is dissolved in alcoholic potash, a salt of the

XSK
composition SC< , known as potassium xanthate, is formed. It

is decomposed by acids, with liberation of carbon disulphide and
alcohol, C2Hf(-OH ; this reaction indicates that the ethyl radical in the
compound is attached to oxygen, and not to sulphur.

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 715

From carbonic acid, by successive replacement of oxygen by sulphur,
a series of acids results :

/OH /OH /SH /SH /SH /SH

OC/ SC< OC< SC< OC< SC<

\OH XOH XOH XOH \SH N3H

carbonic thion-car- thiol-car- thiol-thion- dithiol- thiocar-
acid bonic acid bonic carbonic carbonic bonic acid

acid acid acid

Thiocarbonates give a brown precipitate, CuCS3 with copper salts ;
a red precipitate PbCS3, with lead salts ; and a yellow precipitate,
Ag2CS3, with dilute silver nitrate. These rapidly become black, from
formation of sulphides. Ferric salts give an intense red colour. By
the action of hydrochloric acid on ammonium thiocarbonate, carbon
oxysulphide is evolved.

Carbon disulphide dissolves very readily in solutions of alkali disul-
phides, forming perthiocarbonates : Na2S2 + CS2 = Na2CS4.

Cyanogen, C2N2. — By heating cyanide of silver, Gay-Lussac
(1815) obtained a gas which burns with a peach-blossom coloured
flame. This is cyanogen, C2N2 : 2AgCN = 2Ag + C2N2.

Cyanogen is produced by heating the cyanides of silver, mercury,
and gold, the most convenient being mercuric cyanide, Hg(CN)2,
which is heated to dull redness in a hard glass or steel tube :
Hg(CN)2 = Hg -f- C2N2. A heavy, brown, non-volatile powder is
produced at the same time, called paracyanogen ; it is probably a
polymerised form of cyanogen, (CN)W, and decomposes slowly
into cyanogen at 310°. The gas is evolved at a lower temperature
if mercuric chloride is mixed with the cyanide ; Hg(CN)2 -f- HgCla =
2HgCl -

EXPT. 285. — Heat a little mercuric cyanide in a hard glass tube fitted
with a rubber stopper and glass jet. Ignite the gas at the jet ; it burns
with a characteristic peach-blossom coloured flame. N.B. — Cyanogen
is very poisonous.

The most economical method for the preparation of cyanogen is
to drop a concentrated solution of potassium cyanide into a warm
solution of 2 parts of crystallised copper sulphate dissolved in 4
parts of water. Cupric cyanide, Cu(CN)2, is first formed as a yellow
precipitate, but this quickly decomposes, with evolution of cyanogen
gas, leaving white cuprous cyanide, CuCN : 2CuS04 -}- 4KCN =
2K2S04 + 2CuCN -f~C2N2. If the cuprous cyanide is collected
and treated with ferric chloride solution, the rest of the cyanogen
is evolved : 2CuCN + 2FeCl3 = 2CuCl + 2FeCl2 + C2N2.

Cyanogen is apparently formed directly from its elements when
an arc is struck between carbon poles in nitrogen : 2C -f- N2 ±^
C2N2 ; it can be detected by the spectrum, but is decomposed in

716 INORGANIC ;CHEMISTRY CHAP.

contact with carbon and cannot be separated from the arc-gases.
Cyanides are also produced by the action of nitrogen on carbides
at high temperatures. Thus, if nitrogen is passed over barium
carbide, or an intimate mixture of barium oxide and carbon, at a
red heat, barium cyanide is produced. The compound barium
cyanamide, BaCN2, is first formed: (1) BaO -f- C = Ba -f CO.
(2) Ba + 2C = BaC2. (3) BaC2 + N2 = BaCN2 + C. (4) BaCN2 +
C = BaC2N2. Cyanogen is present in blast-furnace gas in small
quantities. It is an endothermic compound : 2C (graphite) -f- N2 =
C2N2 — 70 kgm. cal.

Properties of cyanogen. — Cyanogen is a colourless gas which is
very soluble in water, and must therefore be collected over mercury.
It has a smell of bitter almonds, and is very poisonous. When
cooled it condenses to a colourless liquid, boiling at — 20-7°, which
freezes below — 35° to a white solid, melting at — 344°. The
density of the gas shows that it has the formula C2N2.

Cyanogen is absorbed by a solution of caustic potash, with the
formation of potassium cyanide, KCN, and potassium cyanate,
KCNO:C2N2 + 2KOH == KCN + KCNO + H2O. With water at 0°
the reaction C2N2 + H20 = HCN -f- HCNO occurs. From the
similarity of these reactions to those with chlorine (p. 368),
and from the fact that all compounds of cyanogen contain the
nnivalent cyanogen group, or radical, CN, the latter is sometimes
written Cy, since it behaves to some degree as an element. In
solutions of cyanides the cyanide ion, CN', is split off : KCN ^±
K' + CN'.

A solution of cyanogen in water decomposes on standing, with
deposition of a brown precipitate of azulmic acid, C4H5N5O : the solu-
tion then contains ammonium oxalate, hydrocyanic acid, HCN, urea,
carbon dioxide, etc.
CN H20 CONH2 H20 COOH H2O COOH

— > (oxamide) ~> I (oxamic acid) -> (oxalic acid)

CN CONH2 CONH2 + NH3 COOH + NH3

A mixture of equal volumes of cyanogen and oxygen explodes
on ignition or with an electric spark, even when carefully dried over
phosphorus pentoxide, with the production of carbon monoxide
and nitrogen : C2N2 4- O2 = 2CO -f- N2 ; with double the volume
of oxygen, carbon dioxide is formed : C2N2 + 202 = 2CO2 -f N2.

The formula of cyanogen may be written as N|OCjN, with
nitrogen tervalent, or as CiN-NiC, with nitrogen quinquevalent.

Hydrocyanic acid, HCN. — When potassium cyanide (p. 793) is
distilled with a mixture of equal volumes of sulphuric acid and
water, the vapour of hydrocyanic acid, HCN, is evolved :

KCN + H2S04 = KHS04 + HCN.
With concentrated sulphuric acid, carbon monoxide is formed in

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 717

large quantities, according to the equation: HCN-|-2H20 =
H-COOH +NH3-H20 + CO + NH3. The gas is dried by a
U-tube of calcium chloride, and passed through a second U-tube
cooled in ice. A colourless liquid, boiling at 26*1° and freezing at
- 15°, collects. This is anhydrous hydrocyanic acid. The vapour
burns with a purple flame in air. The anhydrous acid is also pro-
duced by passing sulphuretted hydrogen over dry mercuric sulphide
heated to 30° in a long glass tube, and condensing the liquid in a
freezing mixture. Hydrocyanic acid is formed when acetylene is
sparked with nitrogen : C2H2 + N2 = 2HCN, or when a mixture
of nitrogen and hydrogen is passed through a carbon

arc

20 + H2 + N2 ^± 2HCN.

Anhydrous hydrocyanic acid is one of the most dangerously
poisonous substances known, and its preparation should be under-
taken only by expert chemists. A dilute solution may be prepared
by distilling potassium ferrocyanide with dilute sulphuric acid
(7 acid + 14 water) ; the 2J per cent, solution is used as a consti-
tuent of remedies for bronchial catarrh, etc., and is called prussic acid.
In this concentration it is also very poisonous.

Hydrocyanic acid is a very weak monobasic acid : its salts
with alkali metals, the cyanides, are hydrolysed in solution. They
show an alkaline reaction, and smell of peach -kernels owing to
the presence of the free acid : KCN -f H20 ^± KOH -f HCN.

The smell of bruised fruit kernels, laurel leaves, and moist bitter
almonds is due to hydrocyanic acid, and it is a curious fact that Scheele,
the discoverer of hydrocyanic acid (1782), did not know of its poisonous
properties : these were first suspected from its formation from the
poisonous bitter almonds, by distillation with water. Ammonia, or
chlorine water, is used as an antidote to the acid, although large doses are
almost instantaneously fatal.

•Hydrocyanic acid in solution appears to exist in two forms, or
tautomeric modifications: H-C |N ^± H-N i C. Two series of com-
pounds with organic radicals, the cyanides, K *C • N, and iso-
cyanides, R-NiC, are known, corresponding with the two hypo-
thetical acids.

Cyanogen chloride, CNC1. — If chlorine is passed into anhydrous
hydrocyanic acid, cyanogen chloride, CN-C1, is formed, which may
be condensed in a freezing mixture to a colourless mobile liquid,
b.-pt. 15-5° (Berthollet, 1787). The liquid, if slightly acidified,
rapid ly polymerises to white, solid, cyanuric chloride, (CNC1)3.
Cyanogen chloride reacts with alkalies, forming a chloride and a
cyanate : CN-C1 + 2KOH = KC1 + KCNO -f H20. It is the
chloride of cyanic acid, HCNO. With ammonia, it forms cyanamide,
CN-NH0.

718 INORGANIC CHEMISTRY CHAP.

Bromine and iodine react with hydrocyanic acid or potassium cyanide
to form white crystalline cyanogen bromide, CN-Br, and cyanogen
• iodide, CN-I, respectively. The latter often occurs as an impurity in
crude iodine. All the halogen compounds of cyanogen are very
poisonous.

Tests for cyanides. — (1) A solution of a cyanide gives with silver
nitrate a white curdy precipitate of silver cyanide, AgCN, soluble in
concentrated nitric acid. (2) To the solution of the cyanide caustic
soda is added, and a few drops of a mixed solution of ferrous
sulphate and ferric chloride: on warming a ferrocyanide is produced:
(a) FeSO4 -f 2KCN = K2SO4 + Fe(CN)2 ; (b) 4KCN + Fe(CN)2 =
K4Fe(CN)6. The dirty-brown precipitate produced is warmed with
concentrated hydrochloric acid, which dissolves the ferric hydroxide
present, and leaves a dark blue residue of Prussian blue (p. 995), formed
by the action of the ferrocyanide on the ferric salt. If only traces of
cyanides are present, a blue or green coloration appears. This test will
detect 1 part of HCN in 50,000 parts of water. (3) The solution is
evaporated to dryness on a water-bath with yellow ammonium sulphide,
when a thiocyanate, e.g., KCNS, is formed : (NH4)2S2 + KCN =
KCNS + (NH4)2S (volatile). The residue is dissolved in water and
ferric chloride solution added : a blood-red coloration of ferric thio-
cyanate, Fe(CNS)3, is formed. This test is very sensitive.

Cyanates. — Potassium, or sodium, cyanide, in a state of fusion,
is a powerful reducing agent : metallic oxides are converted into the
metals, and a cyanate is formed : KCN + PbO = KCNO + Pb.
The cyanate may be extracted with water. When the solution is
acidified, cyanic acid, HCNO, is formed, but is almost completely
decomposed by the water present, with formation of ammonia, and
evolution of carbon dioxide : HCNO -f H20 = NH3 + C02.

Ammonium cyanate, NH4CNO, obtained by mixing concentrated
solutions of potassium cyanate and ammonium chloride, is readily
converted on heating into the isomeric compound urea : NH4-CNO»=
CO(NH2)2. This reaction, discovered by Wohler in 1828, definitely
broke down the hypothetical barrier dividing " inorganic " sub-
stances from " organic " substances, the latter supposed to be pro-
duced only by the agency of the " vital force." The distinction
between inorganic and organic chemistry is now merely one of
convenience. Previously to Wohler's discovery, urea had been
obtained by John Davy from phosgene and ammonia (p. 708), but
he was not aware of the nature of the products of the reaction.

FLAME.

Flame. — A flame is a zone in which chemical combination between
gases is occurring, accompanied by the evolution of heat and light :
briefly, it is composed of glowing gas (Van Helmont, 1648). Trans-

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 719

parent gases, such as nitrogen or oxygen, do not glow when heated
in tubes to a high temperature, nor do burning solids emit flame
unless a vapour is formed. Thus, iron burns in oxygen without a
flame ; carbon burns in air at low temperatures without a flame,
but at high temperatures, when carbon monoxide is formed, the
latter burns with a flame. A flame of pure hydrogen, burning in
dust-free air, does not emit a visible light.

Flame is only produced in chemical reactions when a considerable
amount of energy is liberated, although chemiluminescence, which
may be regarded as a cold flame, can be induced at fairly low tem-
peratures in many cases. Thus, if ether is dropped on a hot iron
plate, so that ignition does not result, a greenish phosphorescent
flame is seen in a dark room.

EXPT. 286. — Thirty c.c. of 30 per cent, hydrogen peroxide are added
to a mixture of 10 c.c. of 10 per cent, pyrogallol solution, 20 c c. of
saturated potassium carbonate solution, and 10 c.c. of commercial
formaldehyde in a dark room. An orange-red glow, accompanied by a
vigorous reaction, is seen. Light of the wave-length emitted is found to
accelerate the reaction, which involves the oxidation of the pyrogallol.

Unless the combustible gas and the supporter of combustion are
mixed before kindling the flame, the latter is hollow, and occupies
only the surface of contact of the two gases. This may be shown
by many experiments.

EXPT. 287. — Depress a piece of new asbestos paper on a Bunsen
flame : a hollow dark ring is formed by the section of the flame. This
may be seen also if a piece of ordinary paper is quickly lowered on to the
flame.

EXPT. 288. — Thrust a match-head quickly inside a Bunsen flame ; or
support the match, head upwards, in the metal tube by a pin stuck
through it, and then kindle the flame. The match-head does not ignite
for a considerable time.

EXPT. 289. — Stretch a piece of fine wire gauze over
a funnel, and place a small heap of gunpowder in the
centre of it (Fig. 347). Pass a rapid stream of hydro-
gen through the funnel, and ignite the gas from above.
The powder remains in the centre of the flame with-
out explosion. If an unlighted match is thrust
quickly through the flame to the powder, there is still
no ignition, but if the flame is slowly turned down, the
match ignites, and the gunpowder explodes.

EXPT. 290. — Insert one end of a glass tube into the FlG 34y _EXperi-

middle portion of a Bunsen flame. Unburnt gas ment to Demon-

, . , , . ,, , strate that a

passes up the tube, and may be kindled at the upper Flame is Hollow.

end.

FIQ. 348.— Air Burn-
ing in Coal Gas.

720 INORGANIC CHEMISTRY CHAP.

The terms combustible and supporter of combustion are purely rela-
tive, and depend simply on which gas is inside and which outside the
flame. This has already been illustrated in the
case of oxygen and hydrogen and hydrogen and
chlorine.

EXPT. 291.— A lamp chimney (Fig. 348) is fitted
with a cork at the lower end, through which pass a
wide straight tube and a narrow tube bent at a right
angle. The top of the chimney is extended by a tin
tube with a flat top pierced by a hole as shown. Coal
gas is passed in through the bent tube, and may be
kindled at the top of the glass, burning with a
luminous flame. At the same time, air is drawn in
through the wide tube, and if a lighted taper is
passed up through this tube into the chimney, the air
ignites and burns in the coal gas with a blue non-
luminous flame. If the top flame is extinguished, and
a taper passed down to the air-flame, it cannot be
kindled, since it is surrounded by an atmosphere of

coal gas, which will not support combustion of

the hydrocarbons of the taper. A jet of air,

however, may be ignited.

If the upper flame is again kindled, and the

supply of coal gas gradually reduced, the upper

flame shrinks and becomes less luminous, whilst

the lower flame increases in size. The increase

in size of the lower flame is due to the circum-
stance that the oxygen has now a more limited

supply of coal gas available, and the combustion

has therefore to extend over a larger area.

Finally the upper flame goes out, partly on

account of the larger proportion of carbon

dioxide in the gas, and partly because a greater

proportion of the coal gas is burnt by the

lower flame.

EXPT. 292. — Arrange the lamp chimney with

a large hole on the top, and two tubes below, as

shown in Fig. 349. Pass the gas from a large

Bunsen burner through the tube A, and kindle a

large flame at the top of the glass. Push the

tube B to the upper part of this flame, and

pass a slow stream of oxygen through it.

Lower B carefully, when it will be seen that

a second flame of oxygen is burning inside the first flame, the oxygen

combining with the unburnt gas in the centre of the large hollow flame.

FIG. 349.— Oxygen Burning
inside a Coal Gas Flame.

XXXIV

OXYGEN COMPOUNDS OF CARBON, ETC.

721

A very accurate account of the structure of flame was given by
Hooke (" Lampas," 1677). He speaks of " that transient shining
body which we call flame " as " nothing but the parts of the oyl
rarined and raised by heat into the form of a vapour or smoak,
the free air that encompasseth this vapour keepeth it into a cylin-
drical form, and by its dissolving property preyeth upon those
parts of it that are outwards . . . producing the light which we
observe ; but those parts which rise from the wick which are in
the middle are not turned to shining flame till they rise towards
the top of the cone, where the free air can reach and so dissolve
them. With the help of a piece of glass [pressed upon the flame]
anyone will plainly perceive that all the middle of the cone of flame
neither shines nor burns, but only the outer superfices thereof that
is contiguous to the free and unsatiated air."

This description refers to a candle or oil-lamp flame. The candle
and lamp consist of a cotton wick, surrounded by the combustible
material. The liquid oil, or the wax melted by heat, rises in this
wick by capillary attraction. The top of the wick becomes incan-
descent, and the fuel is subjected to destructive distillation, the
combustible gases burning with a flame. The action of the wick is
peculiar.

EXPT. 293. — Attempt to kindle a piece of lump-sugar by a taper : the
sugar melts but will not take fire. Now rub a corner of the sugar with
a small quantity of cigarette ash : the sugar can then readily be lighted
at that point and burns with a flame.

In the old tallow candle the wick acquired a deposit of soot, which
required " snuffing " : the wick of the modern paraffin- wax candle
is plaited so that it bends over, and is continuously consumed in the
outer part of the flame. The action
of the wick is probably two-fold :
it presents the combustible material
to the heated zone in a divided
state, owing to its capillary structure,
and it prevents too rapid conduction
of heat away from the heated point
where distillation occurs.

The structure of flame. — A hydrogen
flame burning in air or oxygen con-
sists (Fig. 350) of two cones, an
inner one, A, of unburnt gas, and
an outer one, B, in which the
simple chemical reaction 2H2 + 02 =
2H2O is occurring, with evolution of
heat and light. The flame of ammonia burning in oxygen (p. 548)
consists, however, of three cones, an inner cone, A (Fig. 351), of

3 A

FIG. 350. — Struc-
ture of Hydro-
gen Flame
(two cones).

FIG. 351.— Struc-
ture of Carbon
Bisulphide or
Ammonia
Flame (three
cones).

722

INORGANIC CHEMISTRY

unburnt gas, surmounted by a yellow cone, J5, in which decom-
position of ammonia into its elements is taking place : 2NH3 = N2
-f- 3H2, and an outer pale greenish-yellow cone, (7, in which the
hydrogen burns. The nitrogen largely escapes combustion. A
flame of carbon disulphide vapour in oxygen or air is similar to the

ammonia flame : the
cone B is lilac in
colour,and corresponds
with the reaction :
2CS2 + O2 = 2CO +
4S, whilst the cone C
is deep blue and re-
presents complete
combustion of carbon
monoxide and sulphur.
Hydrocarbon fl a m e s
are more complicated,
and contain four
regions, first defined
by Berzelius. If the
flame of a candle or of
coal gas burning at a

FIG. 352.— Structure of Hydrocarbon Flames. jet (Fig. 352) is exam-

ined, it is found to

consist of (a) the dark inner cone of unburnt gas, or vapour of
partly decomposed wax ; (b) a. yellowish- white, brightly luminous
region, occupying most of the flame ; (c) a small bright blue region
at the base of the flame ; (d) a faintly-luminous outer mantle,
surrounding the flame completely. If the supply of gas is reduced, '
the flame shrinks down, the luminous area 6 grad-
ually disappearing, whilst the region c becomes
continuous, and constitutes an inner cone (Fig.
353). The regions a and d remain.

The luminosity of flame. — The question now
arises : Why are the flames of a candle, coal gas,
and ethylene, for instance, luminous, whilst those
of hydrogen and carbon monoxide are non-
luminous ? There are three theories to account
for the luminosity of flame :

(1) Davy's theory (1816), which ascribed the luminosity to particles
of solid carbon heated to incandescence in the flame. The origin of
the carbon was explained later by Faraday on the incorrect assump-
tion that the hydrogen of hydrocarbons burns preferentially to
carbon, and the latter is deposited.

(2) Frankland's theory (1861), according to which the luminosity
is due to incandescent vapours of dense hydrocarbons in the flame.

Fia. 353. — Small
Hydrocarbon
Flame with Con-
tinuous Blue
Region c.

XXXIV

OXYGEN COMPOUNDS OF CARBON, ETC.

723

FIG. 354.— Principle of
Safety Lamp.

(3) Lewes's theory (1892), which considers the solid carbon in the

flame to be the result of the thermal decomposition of ethylene and

acetylene : C2H4 = C2H2 + H2 = 20 + 2H2.

Davy's investigations on flame. — Sir Humphry Davy in 1815 was

led to the study of flame by an investigation of the causes and

prevention of the disastrous fire-damp ex-
plosions in coal mines, which were prevalent

when open candle flames were used by the

miners. These are caused by the ignition of

mixtures of methane and air (fire-damp) ; or,

as we now know, sometimes by the kindling

of a mixture of very fine coal-clust itself with

air. Davy made the discovery that if a

flame is cooled it is extinguished, and he

recognised that combustible gases have

different ignition points.

EXPT. 294. — Lower a spiral of thick copper

wire over a candle flame : the latter is extinguished. Now heat the

spiral to redness, and repeat the experiment : the flame continues to

burn.

EXPT. 295. — (i) Depress a piece of fine wire gauze over a Bunsen flame.

The flame at first does not pass through the gauze, owing to the cooling
effect caused by conduction of heat through the metal,
and a red-hot ring is seen on the gauze, with a dark
patch in the centre corresponding with the central
unburnt portion of gas in the flame. That unburnt
gas from the central part of the flame is passing
through the gauze may be seen by holding a taper
above the latter. If the experiment is repeated, and
the gauze allowed to remain on the flame a sufficiently
long time, the temperature of the metal rises to the
ignition point, when the gas ignites and burns above
the gauze.

(ii) If a piece of gauze, turned up at the edges, is
held over an unlighted Bunsen burner,, the gas
passing through may be kindled above the gauze, but
the flame does not pass through and light the gas at
the burner. On raising the gauze, the flame flickers
and finally goes out (Fig. 354). This flame, in which

air is mixed with gas before combustion, is blue and non-luminous.

These experiments led Davy to the invention of the safety-lamp,
which consists of an oil lamp having an enclosed cylinder of wire
gauze as a chimney (Fig. 355). If this is taken into a mine where
fire-damp exists, the latter will penetrate inside the gauze and
burn there, but the flame is not propagated to the gas outside,

3 A 2

FIG. 355. — Davy's
Safety lamp.

724

INORGANIC CHEMISTRY

CHAP.

because the heat is conducted away by the metallic gauze. The
gauze may even become red hot from the mixture of gas and air
burning inside, but as the ignition temperature of methane is high,
the flame does not pass through to the gas outside. It has been
found, however, that a draught of air blowing on the lamp may
cause one portion of the gauze to become so hot as to result in
ignition of the fire-damp, and the flame inside may also be blown
mechanically through the gauze by a blast of air, passing at a rate
exceeding 8 ft. per sec., such as is formed on firing a shot in a mine.
With these exceptions the lamp, especially in its improved form,
with a strong glass cylinder below the gauze, which permits of better
illumination, is perfectly safe. The introduction of the safety-
lamp, at first strongly opposed by some miners, has proved a great
boon to workers exposed to fire-damp in the mine. If only a small
amount of fire-damp is present in the air, a flame appears, over the
flame of the lamp, and, from the size of this flame-cap, the amount
of combustible gas in the air may be ascertained.

EXPT. 296. — Lower a lighted Davy lamp into a large beaker into
which some ether has been poured. The interior of the lamp is seen to
be filled with flame, but the ether vapour in the beaker is not ignited.
Davy supposed that the luminosity of a hydrocarbon flame was
due to " the decomposition of a part of the gas towards the interior
of the flame, where the air was in smallest quantity, and the depo-
sition of solid charcoal, which first by its ignition, and afterwards
by its combustion, increases to a high degree the intensity of the
light." The non-luminosity of the flame in the second part of
Expt. 291 was due, according to Davy, to
the carbon particles burning, as fast as pro-
duced, in the oxygen supplied.

Flames known to contain solid particles,
e.g., those of zinc, magnesium, and potassium
in oxygen, are very luminous, and the pres-
ence of solid particles of carbon in luminous
hydrocarbon flames is definitely proved by
the fact that a powerful beam of light is
reflected by such a flame, and the reflected
light is polarised (p. 8). The presence of
carbon particles is also made probable by
the following experiments :

EXPT. 297. — Hold a cold piece of pipeclay
tube in a candle flame. Carbon is deposited
on the lower part only, not on the top.

EXPT. 298. — Clouds of soot evolved from
burning camphor, if admitted to the lower part of a Bunsen flame
through one air-hole by means of a funnel tube (Fig. 356), render the

Fia. 356. — Bunsen Flame
rendered Luminous by
Smoke from Burning
Camphor.

OXYGEN COMPOUNDS OF CARBON, ETC.

725

flame luminous. Powdered charcoal sprinkled into a Bunsen flame also
increases its luminosity.

Faraday accepted Davy's theory, but instead of supposing that
the carbon arose from the decomposition of the gas by heat, he put
forward the theory of the preferential combustion of hydrogen in the
flame, with separation of unburnt carbon, which burnt subse-
quently, e.g. : C2H4 + 02 = 2H2O + 20 ; 2C + 2O2 = 2C02.
Hydrogen was supposed to have a greater affinity for oxygen than
was exhibited by carbon. But Dalton had already shown that if
ethylene is exploded with its own volume of oxygen, all the carbon
is burnt to carbon monoxide, whilst the hydrogen is set free :
C2H4 -f 02 = 2CO +2H2. Faraday's theory is therefore un-
tenable.

EXPT. 299. — The structure of a candle flame is well shown by the
following experiment, due to Faraday. A bent glass siphon is lowered
into the candle flame (Fig. 357). With
the tube just above the wick, dense white
vapours pass over, and condense in the
flask to solid wax : these correspond with
the first process in the flame, the vola-
tilisation of the solid wax by the heat,
which occurs on the wick. This corre-
sponds with the dark central portion of
the flame. On raising the tube into the
bright central portion of the flame, dense
black vapours pass over, which deposit
particles of carbon in the flask. On raising
the tube still further, the black smoke dis-
appears, and steam and carbon dioxide

pass along the siphon. The former condenses to liquid water in the
flask, and if a little lime-water is poured in, the presence of carbon
dioxide is readily proved.

Frankland's theory. — Sir Edward Frankland in 1861 noticed that
the flame of a candle burning on the summit of Mont Blanc emitted
a much feebler light than when it was burnt in the valley at
Chamonix, although the rate of combustion was the same in both
cases. In further experiments he found that a candle flame when
burning under a partially evacuated receiver was much less luminous
than in free air. This had been noticed by Boyle. An alcohol
flame burning in compressed air is luminous. Again, a mixture of
hydrogen and oxygen exploded in a eudiometer burns with a bright
flash, and hydrogen burning in oxygen under 20 atm. pressure
gives a luminous flame. The luminosity of the electric spark
in gases increases with the density of the gas. Luminous flames

FIG. 357.— Faraday's Experiment
to Illustrate the Structure *of a
Candle Flame.

726 INORGANIC CHEMISTRY CHAP.

are known in which solid particles cannot be present, e.g., the flame
of arsenic in oxygen, and of sodium in chlorine. As a result of his
experiments, Frankland suggested that the luminosity of hydrocarbon
flames was not due to the deposition of solid particles of carbon,
as Davy had supposed, but to the presence of dense gaseous hydro-
carbons, which became incandescent. The presence of solid carbon
in flames has, however, definitely been proved, although Frank-
land's theory may apply to flames in which solid matter cannot be
present.

Lewes's theory. — By aspirating and analysing the gases from
different parts of the flame, V. B. Lewes in 1892 found that the
unsaturated hydrocarbons (ethylene and acetylene) disappear only
slowly in the dark portion, but rapidly in the luminous zone. The
proportion of acetylene, however, increases rapidly as the gases
pass up the dark zone, attaining 70 per cent, of the unsaturated
hydrocarbons at the apex of the dark cone. Lewes assumes that
ethylene is decomposed by heat, with the intermediate formation of
'acetylene : C2H4 -> C2H2 + H2 -> 20 + 2H2. The presence of free
hydrogen has been detected in the luminous zone. The carbon is
separated as a fine powder, and the heat of decomposition of the
endothermic acetylene assists in raising the temperature.

The reaction in the bright blue part of the flame appears to be
the same as that in the inner cone of a Bunsen flame (see below) ;
in the outer, faintly visible, cone complete combustion of hydrogen
and carbon monoxide occurs, as in the outer cone of the Bunsen
flame.

The present position of the theory of luminosity of flames may be
summed up in the statement that probably all three causes described
by the theories of Davy, Frankland, and Lewes contribute to the
luminosity.

The Bunsen flame. — If coal gas is mixed with a sufficient supply
of air before combustion, as in the familiar Bunsen burner, it burns
with a non-luminous flame. This now consists only of two cones :

(1) a pale blue inner cone, which becomes green when a large supply
of air is admitted, and. the flame " roars " (as in the Teclu burner) ;

(2) a still paler blue outer cone. The reactions in the inner cone
are different from the purely thermal decompositions taking place
in an ordinary flame, since partial oxidation now occurs, with
formation of carbon monoxide. This burns in the outer cone.

EXPT. 300. — The effect of admixture of air on the flame of a com-
bustible gas may be studied with the apparatus shown in Fig. 358, due
to Smithells. Undiluted carbon monoxide passed in through one of the
lower tubes burns above with a hollow cone of blue flame (a), which is
typical of what Smithells calls a volume flame. If a little air is admitted
the cone becomes shorter, and its inner lining bright blue (b). With

XXXIV

OXYGEN COMPOUNDS OF CARBON, ETC.

727

-EL

JIL

FIG. 358. — Smithells's Experiments on Flames.

continued addition of air, a mixture is finally produced through which a
flame would be propagated without external air, but the flame is kept
on the top of the tube by the speed of the gas current (c). More air
causes the speed
of propagation of
flame through the
mixture to exceed
the speed of the
gas current, and
at this point the
inner cone separ-
ates from the
outer cone in the
flame, and passes
down the tube
(d). At a certain

point the outer cone vanishes, and all the gas now burns in the
inner cone (e). Now the rate of propagation of flame has been dimin-
ished by the excess of air added, and the lower flame is a double cone,
as in the first case. When the rate of inflammation has been reduced
below the rate of flow of gas, the flame again rises to
the top of the tube (/), and burns as a single cone
with a considerable unburnt inner space, typical of a
surface, or film, flame.

EXPT. 301. — The separation of the two cones of a
Bunsen flame is most conveniently effected by means
of Smithells's flame-cone separator (1892). This con-
sists (Fig. 359) of one glass tube sliding inside a wider
tube. A mixture of air and gas (e.g., methane), in
regulated, proportions, is passed into the central tube
through stopcocks at the bottom. The central posi-
tion of the inner tube may be kept by a brass guide
fitted to it or by passing it somewhat loosely through
a cork in the wider tube, as shown. If the quantity
of air supplied is increased, the Bunsen flame burning
at the top separates into two cones, one of which
remains on the outer tube, and the other, which is
the inner cone of the complete flame, passes down
and burns on the top of the narrower tube. By
raising the latter, the inner cone may be joined to the
outer one, and the complete flame raised outside on
the inner tube.

By analysing the interconal gas, drawn off from the space between
the two cones by a side tube shown, it was found to consist of
nitrogen from the air, carbon monoxide, carbon dioxide, steam,

FIG. 359.— Smith-
ells's Flame-cone
Separator.

728 INORGANIC CHEMISTRY CHAP.

and hydrogen. The composition of the mixture was the same if
pure methane, containing no hydrogen, were used, and it is evident
that the reaction taking place in the inner cone of the Bunsen flame
leads to the incomplete burning of the hydrocarbon, with forma-
tion of carbon monoxide and hydrogen (p. 674), and with the excess
of oxygen, when some carbon dioxide is formed, an equilibrium,
CO -f- H2O ^ C02 + H2, is set up between the carbon monoxide,
steam, carbon dioxide, and hydrogen. This is known as the
water-gas equilibrium. The law of mass-action leads to the following
relation between the concentrations :

[CO] X [H20] _ K

[COJ X [HJ

This relation was shown to hold for the water-gas equilibrium by
Dixon in 1884 ; Smithells, and later Haber, find that the same rela-
tion holds for the interconal gases of a flame, and the constant K
has the value corresponding with the temperature of the latter.

The temperatures of flames have been determined in various ways
(e.g., by platinum and platinum-rhodium thermocouples), and the
following values have been found (Fery, 1904) :

Bunsen, fully aerated 1871° Oxy-coal-gas blowpipe 2200°
insufficient air 1712° Oxy -acetylene blowpipe 2420°
Acetylene ... ... 2548° Oxy-acetylene explosion

3000-4000°

Alcohol 1705° [Electric arc 3760°]

Hydrogen, free flame... 1900° [Sun 7800°]

The cause of the non-luminosity of the Bunsen flame has been
attributed to three circumstances : —

(1) Oxidation : Davy's theory, already considered. That this
is at least only a partial explanation follows, however, from the
experiments described below.

(2) Dilution : Blochmann found that not only oxygen, but also
inert gases, such as nitrogen, carbon dioxide, or even steam, will
render the flame of coal gas non-luminous in the Bunsen burner.

EXPT. 302. — Stop up one air-hole at the base of the burner, and
connect the other with an apparatus for generating carbon dioxide.
Light the coal gas, and then gradually admit carbon dioxide : the
flame becomes non -luminous, but consists of only one cone instead
of two, as in the ordinary Bunsen flame.

Lewes states that 1 volume of coal gas requires the following
proportions by volume of inert gases to render it non-luminous :
CO2, 1-26; N8, 2-30; CO, 5-11 ; H2, 12-4. The flame was ren-
dered non -luminous by 0-5 vol. of oxygen, or 2-27 vols. of air.
That the effect cannot be due to cooling entirely is evident from the

XXXIV

OXYGEN COMPOUNDS OF CARBON, ETC.

729

effect of carbon monoxide, which gives a much hotter flame than
coal gas.

(3) Cooling : Wibel showed, however, that cooling the flame
resulted in loss of luminosity.

EXPT. 303. — Bring a cold flat-iron in contact with the flame
of coal gas burning in a fishtail burner. The flame loses its luminosity.

EXPT. 304. — Suspend a platinum crucible in a Bunsen flame which
has been rendered just luminous by adjusting the air-holes, whilst the
crucible is red hot. Now pour cold water in the crucible ; the flame
will be seen to lose its luminosity.

EXPT. 305. — Attach a tube formed by rolling platinum foil round a
glass tube to the top of a Bunsen burner, and light the flame at the
top of the platinum tube. Heat the latter to redness by another Bunsen
flame ; the first flame becomes luminous. This result, A

however, is probably due to the formation of acetylene
on the heated surface as well as to the increased
temperature .of the gas.

The present position of the theory of non-
luminosity is that probably all three causes are
operative.

EXPT. 306. — The principle of the stability of the
Bunsen flame, viz., that the combustible mixture of
gas and air is passed up the tube more rapidly than
the flame is propagated backwards through the mix-
ture, may be illustrated by placing a long wide glass
tube over a large Bunsen burner, and lighting the
flame at the top (Fig. 360). On turning down the
gas, the flame strikes back, i.e., flashes down the tube.

If the gas is turned down very slowly, the inner
cone of the flame may be arrested halfway down the
tube by a ring of copper wire hung inside, as shown.
This prevents the propagation of the flame by cooling the gas below
the ignition temperature.

The detonation wave. — By measuring the speed of the mixture of
gas and air or oxygen necessary to prevent the downward propaga-
tion of a flame in the apparatus described in EXPT. 306, Bunsen
(1867) found that the velocity of propagation of flame in a mixture
of hydrogen and. oxygen was 34 metres per -sec. Later experiments
by Berthelot, Mallard and Le Chatelier, and Dixon showed, however,
that if the explosive mixture is fired at one end of a long tube, the
flame, which at first traverses a short length of the tube with a
velocity comparable with Bunsen 's figure, rapidly increases in speed,
and reaches a maximum, after which it flashes through the gas with

FIG. 360. — Separa-
tion of Cones of
Bunsen Flame.

730

INORGANIC CHEMISTRY

CHAP.

Gas.

8H2
2H
H2

C2
H2

3O,

C12

a constant velocity very much higher than the initial velocity of the
flame. This flame, travelling with the high constant speed, is
called a detonation wave. The velocities of the detonation waves
in various mixtures, determined by Dixon, are given below.

Velocity of

detonation wave

in m. per sec.

3532

2821

1707

2728

1729

In some cases (e.g., C2N2 + O2) the velocity of the detonation wave
is approximately that of the propagation of sound through the burnt
gas heated to the temperature of combustion under the conditions of
experiment : in others (e.g., 2H2 -f O2) it is much higher than this.

The increased violence of the combustion, and the great speed of
propagation of the flame, when the detonating wave has been estab-
lished, may be demonstrated by the following experiments :

EXPT 307. — Fill two tubes with nitric oxide over water, one a large
test-tube, and the other a strong tube 2 in. wide and 5 ft. long, closed
at the ends with rubber bungs. Pour a few c.c. of carbon disulphide
into each, and shake. Take out the stoppers, and ignite the gases
with a taper. The mixture in each burns with a beautiful blue flame,
but whilst that in the test-tube burns quietly
away, the flame in the long tube runs down
noiselessly until it approaches the middle, and
then flashes down quickly, with a peculiar
howling noise. In the long tube the detonation
wave just begins. The lower part of the tube
should be protected with a strong glass screen.

EXPT. 308. — A coil of lead piping, 30 ft. long
and | in. diameter, is fitted at each end with
the ordinary brass coupling sockets used for
gas connections. To one of these is attached,
by a rubber washer, a thin glass test-tube, and
to the other a strong glass tube with firing-
wires sealed through the glass. The glass tube
is fitted into the socket, by Faraday's cement,
and also into a brass stopcock above (Fig. 361).
The coil is filled with a mixture 2CO + O2,

containing a little hydrogen, the test-tube fixed in place, and covered
with a wire gauze cylinder. On passing a spark, the glass tube is
shattered almost at the same instant as the flash is seen in the

FIG. 361.— Velocity of
Detonation Wave.

xxxiv OXYGEN COMPOUNDS OF CARBON, ETC. 731

firing tube. The mixture 2CO + O2 burns in a test-tube without
explosion.

EXERCISES ON CHAPTER XXXIV

1. How is carbon dioxide prepared, and what are its properties ?
How would you demonstrate (a) the gravimetric, (6) the volumetric,
composition of the gas ?

2. How has the atomic weight of carbon been determined ?

3. How .are percarbonates prepared ? How has the composition
of these substances been found ? In what manner may a true per-
carbonate be distinguished from a carbonate containing H2O2 of
crystallisation ?

4. What is the carbon dioxide cycle in Nature ? In what way is the
composition of the atmosphere maintained approximately constant ?

5. How is the amount of carbon dioxide in the air estimated ? What
is the normal proportion, and what effects have an excess of the gas
on health ? Compare the two oxides of carbon, CO and CO2, as regards
their poisonous properties.

6. In what ways do combustion and respiration resemble and differ
from each other ? How is the carbon dioxide content of the blood
regulated ?

7. How is carbon monoxide prepared ? Under what conditions
does the gas combine with (a) oxygen, (6) chlorine ?

8. What reactions are supposed to occur in the burning of carbon ?
How may carbon monoxide be prepared from coke and oxygen ?

9. What are the properties of carbon monoxide ? Describe two
experiments to illustrate the reducing properties of the gas.

10. How is carbon suboxide prepared, and what are its properties ?

11. How are carbon disulphide and carbon oxysulphide prepared?
What are the properties of these substances ? What other sulphides
of carbon have been described ?

12. What are carbonyls ?

13. Describe briefly the manufacture of producer gas, water gas, and
carburetted water gas. How may hydrogen be obtained from water
gas ?

14. Describe the preparation and properties of cyanogen and hydro-
cyanic acid. How may (a) hydrocyanic acid, (b) carbon monoxide,
be prepared from potassium ferrocyanide ?

15. How are the following prepared : (a) potassium cyanate,
(b) sodium thiocarbonate, (c) ammonium thiocyanate? What reaction
takes place when a solution of ammonium cyanate is heated ?

16. Describe experiments to illustrate the following : (a) flames are
hollow, (b) the cause of luminosity of hydrocarbon flames, (c) the
structure of the Bunsen flame.

17. What reactions are supposed to occur in the Bunsen flame ?
What evidence of* these reactions may be offered ?

18. To what causes is the loss of luminosity in the Bunsen flame
ascribed ? What experiments may be performed to support these
theories ?

19. What is a detonation wave ? Describe experiments showing
how such a wave is initiated.

CHAPTER XXXV

BORON AND SILICON

Boron and silicon. — Although boron and silicon belong to two
different groups in the Periodic System, they show many analogies,
and are conveniently studied together. Their general properties,
in relation to the other elements of the groups in which they occur,
will be considered at a later stage (pp. 890, 911).

BORON, B = 10-8.

Boron. — The salt borax, Na2B407,10H20, has been known from
very early times ; it was brought from Tibet, and called tincal.
Borax was used as a flux in metallurgy, and is mentioned by the
Latin Geber. In 1702 Homberg obtained a crystalline substance
by adding oil of vitriol to a solution of borax ; from its medicinal
properties this was known as 'sal sedativum. Baron (1747) showed
that Hombejg's " salt " must possess acidic properties, since, when
it is treated with soda, borax is formed. It was therefore called
boracic acid, oi boric acid, and Lavoisier suggested that it consisted
of oxygen united with a peculiar element, which he called boracium,
or boron. Davy (1807) first obtained boron as an olive-brown
powder by electrolysing moistened boric acid, or by heating fused
boric acid (i.e., boron trioxide, B203) with potassium. The prepara-
tion of boron by the second method was repeated on a larger scale
by Gay-Lussac and Thenard (1808) ; they described the properties
of the element.

Borax. — The greater part of the borax of commerce is prepared
either from the natural borax of Lake Borax, in California, which
contains a little more than one ounce of borax per gallon, or from
minerals, such as :

Colemanite, Ca2B6On,5H20, or 2CaO,3B2O3;5H2O, found in Asia

Minor, and in America ;

Boracite, 2Mg3B8O15,MgCl2, found at Stassfurt ;
Borocakite, CaB4Or,4H2O ;
Boronatrocalcite, Na2B4O7,Ca2B6On,16H2O, found in South

America.

In the preparation of borax, the minerals, such as colemanite,
are ground to a fine powder and boiled with sodium carbonate

732

CH. xxxv BORON AND SILICON 733

solution (15 parts of mineral + 10 parts of Na2CO3 -f- 60 parts of
water) for three hours :

2(2CaO,3B203) -f 3Na2C03 = 3CaC03 + CaO + 3Na2B407.

The solution is filtered, and allowed to crystallise for three days in
vats. The borax is drained, broken up, and packed in kegs.

Borax forms two important hydrates : octahedral borax,
Na2B4O7,5H20, is obtained by crystallisation from a warm solution,
above 35-5° ; at lower temperatures the salt deposits as common,
monoclinic borax, Na2B4O7,10H2O. The crystals, and powder,
swell up considerably on heating, forming anhydrous borax, which
fuses at a higher temperature to a transparent glass. Borax is
slightly hydrolysed in solution, and since boric acid, H3BO3, is a
very weak acid, the solution is alkaline : Na2B407 + 3H2O ^
2NaBO2 + 2H3B03 (concentrated solutions) ; NaB02 + 2H2O =±
NaOH -f H3BO3 (dilute solutions). Borax is used in laundering
for imparting a gloss to linen in ironing, and (on account of the pro-
perties of boric acid) as an antiseptic. Fused borax readily dis-
solves metallic oxides, often producing charac-
teristic colours (borax-bead reactions : CuO, blue ;
Cu2O, red ; Cr2O3, green ; Mn02, violet ; CoO,
deep blue ; NiO, yellowish -brown ; FeO, green ;
Fe2O3, brown). Borax is used in preparing
glazes, as a flux in soldering, and in making
optical and hard glass.

Boric acid, H3B03. — Boric acid is produced FIQ. 3627— crystal
from borax by treating it with a mineral acid. of Boric Acid-

It is sparingly soluble in cold water, but
dissolves fairly easily in hot water. One hundred c.c. of
water dissolve 1-95 gm. at 0°, 2-92 gm. at 12°, and 16-82 gm. at
100°.

EXPT. 309. — To a hot saturated solution of borax add concentrated
hydrochloric acid till the solution is strongly acid to litmus. On cool-
ing, scaly, six-sided, triclinic crystals of boric acid (Fig. 362) sepa-
rate : Na2B4O7 + 2HCl + 5H2O = 2NaCl + 4H3BO3. Wash the crystals
with cold water, and recrystallise them from hot water.

In the volcanic regions of Tuscany, jets of steam, called suffioni,
escape from the ground, and are surrounded by lagoons ; these jets
contain steam, nitrogen, ammonia, and traces of boric acid, which
is volatile in steam. The boric acid of suffioni has probably been
produced by the action of superheated water on boron nitride :
BN + 3H20 = H3B03 + NH3. In the recovery of the boric acid,
a basin is built around two or three of the suffioni, and the steam
is condensed in water. The liquid is concentrated by the heat of
the steam ; it passes through successive basins on a sloping hillside

734

INORGANIC CHEMISTRY

(Fig. 363), and becomes enriched in boric acid. The liquid, con-
taining about 2 per cent, of the acid, is then concentrated in flat
lead pans by the heat of the steam, and the crystals of boric acid
separating are recrystallised and dried.

Ordinary boric acid, or orthoboric acid, H3BO3, forms soft, silky,
white crystals with a greasy feel. On heating to 100°, these lose
water and form metaboric acid, HB02. At 140°, pyroboric acid,
H2B4O7,' is said to be formed ; at a red heat the whole of the water
is lost, with formation of boric anhydride, or boron trioxide, B203,
which softens to a hygroscopic, glassy mass at a red heat :

4H3B03 = 4HBO2 + 4H20 = H2B4O7 + 5H20 = 2B203 + 6H20.

FIG. 363. — Boric Acid Lagoons.

The constitutional formulae of the acids may be written :
Orthoboric acid, B(OH)3 ; metaboric acid, O:B(OH) ; pyroboric acid,

(OH)B

No/

B— O— B<

>B(OH).

Orthoborates are infrequent : magnesium borate, Mg3(B03)2, and
ethyl borate, B(OC2H5)3, are best known. Metaborates are the
most stable, and pyroborates are also stable. Borax, or sodium
pyroborate, Na2B407,10H20, is formed by adding a solution of
caustic soda, or sodium carbonate, to boric acid : since it contains
twice as much boric anhydride, B203, as the normal salt, it is often
called a diborate : Na20,2B203. Metallic borates, usually meta-
borates, are precipitated by adding a solution of borax to the

xxxv BORON AND SILICON 735

metallic salts dissolved in water : Na2B4O7 -f BaCl2 -f 3H20 =^
Ba(B02)2 (barium metaborate) + 2H3B03 + 2NaCl. Metaborates
are also formed in the borax-bead reaction : Na2B4O7 -f- CuO =
Cu(B02)2 + 2NaB02.

Boron trioxide shows feebly basic properties as well as being the
anhydride of the weak boric acid. Boric acid combines with
sulphur trioxide, forming boron hydrogen sulphate, B(HS04)3, and
with phosphoric acid to produce boron phosphate, BP04, insoluble
in water and dilute acids, but soluble in alkalies. In this respect,
boron resembles aluminium (p. 896).

Boric acid is a very weak acid. It turns litmus a wine-red colour,
but has no action on methyl-orange. It is weaker than carbonic
acid, or even hydrogen sulphide, as is seen from the following
table of the fractions ionised in 0-1 normal solutions at 18° :

a a

Hydrochloric acid ... 0-92 Carbonic acid (H-HCO3) 0-0017

Sulphuric acid 0-61 Hydrogen sulphide (H-HS) 0-0007

Acetic acid 0-013 Boric acid (H-H2BO3) 0-0001

Hydrocyanic acid ... 0-0001

Boric acid ionises as a monobasic acid, and may be titrated with
caustic soda after addition of excess of glycerin, with phenol-
phthalein as indicator : H3B03 -f NaOH = NaB02 + 2H2O. Since
the acid has no action on methyl-orange, a solution of borax may
be titrated with this indicator as if it were a solution of caustic
soda : Na2B407 -f 2HC1 + 5H2O = 2NaCl + 4H3B03.

Boron. — The element boron may be obtained by heating boron
trioxide with potassium (Davy) : B203 + 6K = 2B -f 3K20. It
is more conveniently prepared by heating potassium borofluoride
(q.v.) with potassium : KBF4 + 3K = 4KF -f B, but the most
convenient process is to heat boron trioxide with magnesium :
B2O3 -f 3Mg = 2B -f 3MgO. The chestnut brown powder left on
treating the mass with hydrochloric acid may be purified by treat-
ment with hydrofluoric acid and fusion with B203 in a stream of
hydrogen (Moissan, 1902).

EXPT. 310. — Heat 5 gm. of magnesium powder with 15 gm. of
powdered boron trioxide in a covered crucible. When the violent
reaction ceases, coo], and place the crucible in a beaker containing
diluted hydrochloric acid (1 : 2). Filter and wash. In the later stages
of the washing, observe that the boron begins to pass through the filter-
paper in the form of a yellowish-brown colloidal solution, from which it
is precipitated by acids and salts. Dry the boron in a steam oven.

Amorphous boron so prepared is a brown powder, sp. gr. 2 '45 ;
it is unaltered in air at the ordinary temperature, but smoulders
at about 700°, with formation of the trioxide and boron nitride,

736 INORGANIC CHEMISTRY CHAP.

BN. These produce a superficial coating over the boron and pre-
vent complete reaction. Boron displaces carbon and silicon from
their oxides on heating : 3SiO2 + 4B = 2B203 + 3Si.

Moissan's boron, prepared as above, always contains oxygen, and is
probably a solid solution of boron suboxide, B4O3, in boron. Weintraub
(1909) states that pure boron is insoluble in 40 per cent, nitric acid,
which dissolves a considerable proportion of Moissan's boron, leaving a
residue of pure boron. Pure boron is obtained by striking an alternating
current arc in a mixture of hydrogen and boron trichloride vapour,
between water-cooled copper electrodes in a glass globe. The boron
powder collecting on the electrodes fuses to globules, which drop off
(Pring and Fielding, 1910). As so prepared, boron forms a black,
very hard, solid with a conchoidal fracture, melting at 2200°, but
volatilising appreciably at 1600°. It may be strongly heated in air
without oxidation, and is only very slowly attacked by concentrated
nitric acid. It thus differs in properties from Moissan's boron.

Boron is one of the few elements which combine directly with
nitrogen (p. 540) : the nitride is also produced by heating borax
with ammonium chloride :

Na2B4O7 + 4NH4C1 - 4BN + 2NaCl + 2HC1 + 7H2O.
When boron is heated in nitric oxide it burns : 5B + 3NO =
B203 + 3BN. Boron nitride, BN, is a white infusible powder,
unchanged by mineral acids, solutions of alkalies, or chlorine at a
red heat. It is decomposed by fusion with potash, when heated
in steam : 2BN + 3H2O = B2O3 + 2NH3, or (slowly) by hydro-
fluoric acid : BN + 4HF — NH4BF4. When fused with potassium
carbonate, it forms potassium cyanate : BN -}- K2C03 =
KB02 + KCNO.

Boron forms the carbides, BC p-nd B6C, on heating with carbon in the
electric furnace, and a sulphide, B2S3, by direct combination at a white
heat, or by heating B2O3 + carbon in the .vapour of CS2. The sulphide
is hydrolysed by water : B2S3 + 3H2O = B2O3 + 3H2S. With H2S,
metathioboric acid, H2B2S4, is formed. B2S5 is also known, formed
from BI3 and S dissolved in CS2.

Crystalline boron was obtained by Deville and Wohler (1856) by
fusing boron with aluminium at 1300°. On cooling, crystals
formed on the surface of the aluminium. The metal may be dis-
solved in hydrochloric acid, leaving crystals of adamantine boron —
some clear and colourless, others brown, but all having the crystal-
line form of the diamond. Crystalline boron is very resistant to
heat or acids, but dissolves in fused alkalies. The crystals always
contain about 4 per cent, of carbon and up to 7 per cent, of alu-
minium, and are usually regarded as a definite compound, A1B12,
or B48C2A13. Graphite-like laminae of A1B]2 are also formed in
Wohler 's process.

xxxv BORON AND SILICON 737

Boron hydrides. — If equal weights of boron trioxide and magne-
sium powder are heated, magnesium boride appears to be formed,
since the residue, when treated with hydrochloric acid, evolves
hydrogen with has a peculiar smell, and burns with a green-edged
flame (Francis Jones, 1879). Ramsay and Hatfield (1901) showed
that the gas, which was supposed to contain a tri-hydride BH3,
contains several hydrides, which may be condensed out in liquid
air, but no BH3. The investigations of Stock and his pupils since
1912 have shown that probably ten hydrides of boron exist ; a gas,
B2H6, two volatile liquids, B4H10 and B6H12, and several solid
hydrides of doubtful formulae. BH3 does not exist.

The liquid condensed out of the gas from magnesium boride and
hydrochloric acid by cooling in liquid air is a mixture of the two hydrides
B4H10 and B6H12, which can be separated by fractionation, the latter
being less volatile. At the ordinary temperature, these hydrides are
colourless liquids, boiling at 16° and about 100°, respectively. By
heating B4H10, a colourless gas, B2H6, which forms a liquid boiling at
— 87° and very stable when pure, is obtained. This reacts with water :
B2H6 + 6H2O = 2H3BO3 + 6H2. It probably contains quadrivalent
boron : H3B'BH3. On heating B2H6, several solid hydrides are formed.
One of these, B10H14, is volatile in vacuo, and soluble in alcohol, ether,
and benzene. A colourless solid, possibly B12H, is non-volatile but
soluble in carbon disulphide, whilst a yellow solid, possibly B6H]4, is
non-volatile and insoluble in that solvent. By the action of B2H6 and
B4H10 on solutions of alkalies, unstable hypoborates, RO-BH3, are
formed. By the action of chlorine on B2H6, the compound B2H6C1X is
obtained.

Halogen compounds of boron. — The following halogen compounds
of boron are known :

BF3 ; colourless gas, condensing to colourless, mobile liquid, m.-pt.

- 127°, b.-pt. - 101°.

BC13 ; colourless, mobile liquid, m.-pt. — 104°, b.-pt. 12-5°. sp. gr. 1-4.
BBr3 ; colourless, viscous liquid, m.-pt. — 46°, b.-pt. 99°.
BI3 ; white, leafy crystals, m.-pt. 43°, b.-pt. 210°.

Boron fluoride, BF3, is obtained by the spontaneous combustion
of boron in fluorine, or by heating a mixture of fluorspar, boron
trioxide, and concentrated sulphuric acid in a lead retort :

B203 + 3CaF2 + 3H2S04 = 2BF3 + 3CaSO4 + 3H20.

The gas is collected over mercury. It fumes strongly in moist air,
and when passed into water gives a precipitate of boric acid ; this
redissolves if more gas is passed through, and the solution then
contains fluoboric acid, HBF4 : 4BF3 + 3H20 = B(OH)3 + 3HBF4.
The solution on distillation gives a strongly acid liquid of composi-
tion BF3,2H20 ; in concentrated solutions BF3 and HF are also
present. The acid forms salts, borofluorides, e.g., KBF4. BF3 readily
combines with ammonia, giving a white solid, BF3,NH3, and liquids

3 B

738 INORGANIC CHEMISTRY CHAP.

supposed to be BF3,2NH3 and BF3,3NH3. Borofluorides are formed
in solution from boric acid and acid fluorides :

H3B03 + 2NaHF2 = NaBF4 + NaOH + 2H20.

Boron chloride, BC13, is obtained by burning amorphous boron in
chlorine, or by passing chlorine over a strongly-heated mixture of
boron trioxide and charcoal : B203 + 3C + 3C12 = 2BC13 + SCO.
It is condensed in a freezing mixture.

Chlorine has an affinity for boron, and carbon for oxygen, but neither
element alone can effect the decomposition of boron trioxide. The
united action of the two affinities can, however, resolve the oxide.

Boron trichloride is also produced by heating B2O3 with phosphorus
pentachloride in a sealed tube at 150°: B2O3 + 3PC16 = 2BC13 + 3POC13.

The liquid is freed from chlorine by distillation over mercury.
It fumes strongly in moist air, and is immediately hydrolysed by
water : BC13 + 3H20 = B(OH)3 + 3HC1 ; the reaction is not
reversible (cf. p. 450). When passed into liquid ammonia at
— 23° it forms boron amide, B(NH2)3 ; at 0° B2(NH)3, boron imide,
is formed.

The bromide, BBr3, is obtained by similar methods to the chloride ;
the iodide, BI3, is formed by passing BC13 and HI through a heated tube.

Perborates. — If a mixture of boric acid and sodium peroxide is
added to ice-cold water, a perborate, Na2B408,10H20, is produced. On
treating with hydrochloric acid, this forms a meta-perborate,
NaB03,4H20. A salt, Na2B4On, is formed from H202, Na2O2, and
pyroboric acid at 0°. These salts are stable in the solid state, but
they decompose in solution with formation of hydrogen peroxide.
They liberate iodine from iodides, and decolorise permanganate.
Perborates are used in bleaching, and as antiseptics. The salts
2KBO3,H20 and 2KB03,H202 are known. Perboric acid is given

the formula O:BO-OH, or possibly HO'

Tests for boric acid. — If a solution of a borate is acidified with
hydrochloric acid, and a piece of turmeric paper dipped into the
solution and dried, a brownish-red colour is produced, similar to
that formed by alkalies. If the paper is now moistened with an
alkali, it turns greenish-black.

Ethyl borate, B(OC2H5)3, is formed when a borate is distilled with
alcohol and concentrated sulphuric acid : B(OH)3 -f- 3C2H5OH ^
B(OC2H5)3 + 3H20. The vapour of this compound burns with a
green flame.

EXPT. 31 1. — Add a little borax, and then concentrated sulphuric acid,
to alcohol in a dish. Stir well and ignite. The flame is tinged green,

XXXV

BORON AND SILICON

739

especially if blown out and rekindled. Since copper and barium salts
also colour the alcohol flame green, the test is most satisfactorily made
by heating the mixture in a small flask fitted with a glass jet (Fig. 364),
and burning the vapours after admixture with air in a wider tube to
destroy the luminosity of the flame (due to ether, (C2H5)2O, also formed).
Since boric acid interferes in qualitative analysis with the separa-
tion of the metals in Groups III, IV, and V (p. 630), and mag-
nesium, it is removed, if its presence has been detected, by repeated
evaporation of the solution with dilute hydrochloric acid. The.
boric acid is volatile in steam, and is slowly but completely eliminated.
If the acid is not removed, insoluble borates, e.g., calcium borate,
Ca(B02)2, are precipitated by ammonia in Group III.

Lower oxides of boron. — The oxide B4O3 is sup-
posed to be contained in Moissan's amorphous
boron. By decomposing magnesium boride with
cold water, evaporating the filtered solution in
vacuo, and heating, the oxide B2O2 is obtained
(Travers, 1914). B4O5 is obtained by adding
magnesium boride to water : Mg3B2 -J- 6H2O =
Mg3B2(OH)6 + 3H2, and treating the compound
Mg3B2(OH)6 with ammonia for several days in an
atmosphere of hydrogen. The filtrate is evaporated
to dryness in vacuo, when a pale brown solid,
B4O5, is left. The magnesium compound derived
from B2(OH2)G has been called a borohydrate, by
analogy with carbohydrates. The solution also
contains -small quantities of compounds which
evolve hydrogen with acids, possibly borohydrates,
of the formulae H6B2O2 and H6B2O3Mg, in which
boron is quinquevalent. (Travers, Ray, and Gupta,
1912.)

SILICON, Si = 28-1.

Silica. — Next to oxygen, silicon is the most
abundant element in the crust of the earth
(p. 32) : it occurs in combination with oxygen
as silicon dioxide, or silica, Si02, varieties of which are quartz,
sand, flint, etc. Silica is also the acidic constituent of the very
abundant silicate rocks. Granite and similar primitive rocks contain
from 20 to 30 per cent, of silicon. Silica was at first regarded as an
" earth," analogous to lime and alumina, but its acidic character
was pointed out by Otto Tachenius in 1668 : it is insoluble in acids,
but dissolves in potash, forming a solution of a silicate, formerly
known as liquor of flints. Tachenius also observed that acids
differ in strength ; one acid is displaced from its compounds by a
stronger acid. The acidic character of silica explains the formation
of slags in metallurgical operations. These are glassy or stony
masses formed in smelting ores containing silica or silicates, to which

3 B 2

FIG. 364. — Green
Flame of Ethyl
Borate.

740 INORGANIC CHEMISTRY CHAP.

lime has been added, and consist principally of the silicates of
calcium and aluminium.

Lavoisier, who included silica among the earths, expressed the
opinion that the latter " must soon cease to be considered as simple
bodies," and are probably " compounds consisting of simple sub-
stances, perhaps metallic, oxydated to a certain degree." Gay-
Lussac and Thenard in 1801 obtained a brown amorphous powder
on passing the vapour of silicon chloride over heated potassium ;
this was silicon, the element of which silica is the oxide, but its true
character was not elucidated by the French chemists. In 1823
Berzelius prepared silicon by heating potassium silicofluoride with
potassium : K2SiF6 -f 4K = 6KF + Si. He considered it to be a
metal — silicium — whereas Davy, from its analogy with carbon,
regarded it, as a non-metal. In most of its properties silicon belongs
to the group of non-metallic elements, although it forms alloys with
metals, such as copper and iron. It differs from carbon, which also
forms alloys, by giving a solid, difficultly-fusible dioxide, SiO2,
which is the chemical analogue of carbon dioxide, C02. The
remaining compounds of silicon, however, resemble more closely
those of carbon :

Carbon tetrachloride, CC14, b.-pt. 76°; silicon tetrachloride, SiCl4,

b.-pt. 57°.

Chloroform CHC13, b.-pt. 60° ; silicon chloroform, SiHCl3, b.-pt. 34°.

The great difference in physical properties between silica and carbon
dioxide would therefore seem to be due rather to some peculiarity of
silica itself than to the element silicon. Probably it has its foundation
in the highly polymerised nature of the molecule of silica, (SiO2)n,
since substances of high molecular weight usually have high boiling
points (c/. H2S and H2O, p. 482). This is confirmed by the examination
of quartz by the X-rays (p. 1018).

The forms of silica. — Silica occurs both crystallised and amorphous .
Three main crystalline forms have been described, viz., quartz,
tridymite, and cristobalite, although two forms of each are said to
exist having definite transition points (Fenner, 1912) :

575°

a- quartz (tetartohedral hexagonal) ^± /3- quartz (hemihedral hex-
agonal).

870° ±10°

/3-quartz ^± /3- tridymite (holohedral hexagonal),

1470° ± 10°

/3-tridymite ^ /3-cristobalite (cubic).

By rapidly cooling /3-tridymite and /3-cristobalite, they pass at the
following temperatures into metastable forms with a lower optical
symmetry (p. 434) :

115° -120°

/3-tridymite ^ a-tridymite (biaxial, perhaps orthorhombic),

180° -270°

/3- cristobalite ^ a-cristobalite (biaxial).

BORON AND SILICON

741

Silica occurs not; only in the mineral kingdom, but also as a
constituent of vegetable and animal organisms. The straw of
cereals and the bamboo cane contain it in fairly large quantities :
the common weed " horse-tail " leaves on combustion a siliceous
skeleton. The feathers of some birds contain 40 per cent, of silica,
which also occurs in sponges, and vast deposits of almost pure silica
are found at Hanover, near Berlin, and in other localities, in the
form of kieselguhr, which consists of the siliceous skeletons of extinct
diatoms. This material, being very porous, is used to absorb nitro-
glycerin in the preparation of dynamite, and in lagging steam
pipes^to retard loss of heat.

Superheated water in the interior of the earth, especially if
alkaline, dissolves silica : the
latter occurs in many spring
waters, in hot-springs
(Black, 1794), and particu-
larly in the boiling water
of geysers, such as the Great
Geyser of Iceland, the Hot
Springs of New Zealand,
and the Mammoth Springs
of Yellowstone Park, U.S.A.
The dissolved silica is de-
posited in the hydrated
form at the mouth of the
geyser, as sinter. It may
also pass into the pores of
wood, etc., in the earth, pro-
ducing petrifaction.

Quartz. — Quartz (sp. gr.
2 • 66 ) , or rock-crystal,

formerly believed to be "a
hard form of ice," occurs
sometimes in clear, colourless

crystals (Figs. 365-6) used for the preparation of spectacle lenses
(" pebbles "), prisms, and optical apparatus, but is more fre-
quently found in opaque (" milky ") or coloured masses (" smoky-
quartz," " cairngorm ") : coloured varieties of quartz (e.g., purple,
in amethysts) are used as gems. Sand consists of quartz, which
remains unchanged after the disintegration, or " weathering," of
rocks, and has been crushed during its movement by water.

The purest forms of sand are white (" Calais sand ") ; yellow
sand is coloured by ferric oxide, much of which may be dissolved by
boiling with hydrochloric acid. Clay may also be present. Sand-
stone consists of sand grains cemented together, with oxide of iron,
or other materials.

FIG. 365.— Crystals of Quartz.

742

INORGANIC CHEMISTRY

CHAP.

" Singing sand," which emits a peculiar squeaking note when pressed,
consists of rounded grains of nearly uniform size. It occurs in patches
along with ordinary sand in various localities — e.g., near Poole.

The crystalline form of quartz is somewhat complicated ; it is appa-

FIG. 366. — Crystals of Quartz (British Museum).

rently that of the hexagonal prism, terminated by the hexagonal pyramid,
but really belongs to the trigonal system of symmetry, and possesses
optical activity of a peculiar kind. Some crystals exhibit hemihedral

faces (p. 440) inclined to the right,
others to the left, so that one type
of crystal is the mirror-image of the
other (Fig. 367). Such pairs of
crystals are known as enantio-
morphs, and they are said to show
the crystallographic phenomenon
of enantiomorphism. This two-
sided character of the outer form
has its counterpart in the internal
structure of the crystal, as ex-
hibited by its optical properties :
right-handed, or dextrogyrous,
quartz crystals rotate the plane
of polarised light to the right,
left-handed, or Icevogyrous quartz
crystals rotate the plane of polar -
The rotation is proportional to the thickness of

FIG. 367.-Enantiomorphous Crystals of
Quartz.

isation to the left.
crystal traversed.

Tridymite. — Tridymite (sp. gr. 2-30) occurs more rarely than
quartz, in minute crystals, usually in the form of six-sided plates

xxxv BORON AND SILICON 743

(Fig. 368), in cavities in the trachytic rocks of Mexico, and Stenzel-

borg. It belongs to the triclinic system.

Quartz and tridymite appear to have been deposited from solution.

If hydrated silica (p. 744) is heated with a solution of soluble glass

(sodium silicate) in a sealed glass tube,
small crystals of quartz are formed. A
solution of soluble glass alone dissolves
part of the glass tube, and on cooling

silica is deposited ; above 180° quartz is
FIG. 368— Crystalline Form of

Tridymite. formed, at lower temperatures tridymite,

and at the ordinary temperature amor-
phous silica. Larger crystals of quartz are produced by the prolonged
heating at 250°, in a sealed tube, of a 10 per cent, solution of colloidal
silica.

Cristobalite. — This crystalline variety, discovered by Schwarz
(1912), is obtained by heating powdered amorphous (fused) quartz
to 1500°. It has a specific gravity of 2-519.

Amorphous silica. — All the varieties of silica fuse in the oxy-
hydrogen blowpipe at about 1625°, and boil in the electric furnace
at 1700-1750°. They become plastic before fusion, and may be
worked and blown like glass, or drawn into thread. The amor-
phous, vitreous product, called quartz glass, has a very small coeffi-
cient of expansion (cubical coefficient = 10 ~7), and may therefore
be heated to redness and quenched in cold water without fracture.
It is transparent to the ultra-violet rays, whilst ordinary glass is
opaque.

Besides the transparent silica obtained by fusion, a translucent
variety, known as vitreosil, is manufactured by fritting sand with
an electrically-heated carbon rod or plate, previously wrapped in
paper, which when carbonised prevents the fused silica from sticking
to the carbon heater.

Amorphous silica occurs in Nature in a variety of forms. Masses
of quartz are apparently amorphous, and break with a conchoidal
fracture, but probably have a fine (cry ptocry stalling) crystalline
structure. Mixtures of amorphous silica with quartz or tridymite
occur as chalcedony, which is translucent and yellow (sp. gr. 2-3) ;
other varieties are the gems carnelian (red), sard (brown-red),
chrysoprase (apple-green), onyx, and sardonyx (red). Common flint
occurs in rounded nodules in chalk (" chert "), coloured yellow,
grey or black by oxide of iron. It is very hard, and splits with
a conchoidal fracture, giving sharp edges — hence its use in the
"Stone Age." The opal (sp. gr. 2-2) contains 2-13 per cent, of
water, and has, like other amorphous varieties, apparently been
formed by the drying of colloidal silica (q.v.). The noble or gem
opal shows brilliant colours by the interference of light in

744 INORGANIC CHEMISTRY CHAP.

thin layers. Waxy opal is found in large quantities in
Queensland.

Mixtures of the above forms with crystalline quartz and tridymite
occur. Crystals of tridymite are often left on treating opal with
caustic potash. Agates, used in making mortars, are mixtures of
opal or chalcedony with quartz or tridymite, and have a banded
structure, which seems to indicate that they have been deposited
in layers from water on the sides of a " pipe." The cat's eye con-
sists of crystals of quartz enclosing fibres of asbestos. Jasper is
opal deposited in layers of various colours.

Pure silica occurs as transparent rock-crystal, or may be obtained
in the amorphous form from mineral silicates by fusing the finely-
powdered mineral with excess of a mixture of potassium and sodium
carbonates in a platinum crucible until evolution of carbon dioxide
ceases. Alkali silicates (e.g., sodium metasilicate, Na2Si03) are
formed : Na2C03 + SiO2 = Na2Si03 -j- C02. Commercial sodium
silicate has approximately the composition Na20,4SiO2. The residue
on cooling is powdered and boiled with hydrochloric acid, which
dissolves impurities, such as oxide of iron, and precipitates gelatinous
silica, a hydrated form. The whole is evaporated to dryness on a
water-bath. The silica then becomes granular and quite insoluble
in water. It is washed with boiling hydrochloric acid until quite
free from iron, then with boiling water till free from acid and alkali-
chlorides, and is finally heated to redness in a platinum dish. It
forms an impalpable white powder, insoluble in water and all acids
except phosphoric and hydrofluoric. It dissolves in hot concen:
trated caustic alkalies.

The above process serves for the detection and estimation of silica in
minerals, and manufactured products. A simple qualitative test is to
heat a fragment of the mineral in a microcosrnic bead (p. 633). All
metallic oxides dissolve in the sodium metaphosphate, and a skeleton
of silica is left floating in the bead. A sodium carbonate bead dissolves
silica with effervescence, and remains clear on cooling : sodium silicate
is formed.

At high temperatures, silica, being a practically non- volatile
acidic oxide, is able to displace volatile acids from their salts. If
heated with sodium sulphate it drives out the volatile sulphur
trioxide : Na2S04 -f Si02 = Na2Si03 + S03. It is, however, rela-
tively inert and refractory, and is used for making refractory bricks
(ganister, Dinas brick, etc.) for furnace-linings. For this purpose
pure sand, or crushed quartz-rock, is mixed with a little lime and
clay, and old broken firebrick (" grog ") ; the mass is moistened,
moulded, and burnt.

Silicic acids. — Gelatinous silica, freshly precipitated by the
addition of acids to solutions of sodium or potassium silicates, is
appreciably soluble in water, alkali, sodium carbonate, and acids.

xxxv BORON AND SILICON 745

When dried in the air, it retains about 16 per cent, of water, corre-
sponding roughly with the formula, SiO2,H2O or H2Si03, of meta-
silicic acid. At 100°, 13 per cent, of water remains, and the silica
is then insoluble. On further heating, water is gradually lost, but
the vapour -pressure curve shows no breaks indicative of hydrates
(p. 204). If water- vapour is readmitted to the partially dehydrated
mass, it is reabsorbed, but the pressure is higher than in the corre-
sponding part of the dehydration curve. At about 500° all the water
is lost. The hydrated form of silica precipitated when silicon
fluoride or chloride is decomposed by water (p. 752) is often assumed
to be orthosilicic acid, Si(OH)4 ; it has this composition when washed
rapidly with benzene and ether and dried between filter-paper at the
atmospheric temperature, but the existence of Si(OH)4 is doubtful.
The relations between the ortho- and meta-acids and the anhydride
would be as follows :

- H2o - H2o

Si(OH)4 -> SiO(OH)2 -> Si02

or Si«X5 -> 0=Si -> ==.

\ v \OH

X)H

Colloidal silica. — If a dilute solution of sodium silicate is
poured slowly, with stirring, into an excess of dilute hydrochloric
acid, no precipitation of silica occurs, although the reaction
Na2Si03 + 2HC1 = 2NaCl + (SiO2 + H2O) has taken place, as may
be shown by the diminution in electrical conductivity consequent
upon the disappearance of the hydrogen ions. If the liquid be
poured on a dialyser (p. 314), the sodium and chloride ions diffuse
out, leaving a clear colloidal solution, or hydrosol, of silicic acid
(p. 316). This was discovered by Graham in 1861. The colloidal
solution may be concentrated by boiling in a flask to a certain
extent, and by further evaporation over sulphuric acid until it
contains 14 per cent, of Si02 ; it is then a clear, tasteless liquid with a
feebly acid reaction. It is readily coagulated to a bluish- white, nearly
transparent, jelly, the hydrogel of silicic acid. This, when washed
with 90 per cent, alcohol, has approximately the composition
H2Si03. The hydrosol is more stable if small amounts of hydro-
chloric acid or caustic soda are added, but is at once coagulated
by sodium carbonate or phosphate.

The silicates.— Although silicon does not form such a large number
of compounds as the element carbon, the silicates enter into the
composition of an extensive series of rock-forming minerals, the
formulae of which are often rather complex. Most silicates, however,
may be regarded as salts of six- hypothetical silicic acids, viz., ortho-

746 INORGANIC CHEMISTRY CHAP.

silicic acid, Si(OH)4, and acids produced from one or more mole-
cules of this by elimination of water.

Many mineral silicates have been prepared artificially.

1. H4SiO4 : orthosilicic acid.

2. H4SiO4 - H2O = H2SiO3 : metasilicic acid.

3. 2H4SiO4 - H2O = H6Si2O7(2SiO2,3H2O) : orthodisilicic acid.

4. 2H4SiO4 - 3H2O = H2Si2O5(2SiO2,H2O) : metadisilicic acid.

5. 3H4SiO4 - 2H2O = H8Si3O10(3SiO2,4H2O) : orthotrisilicic acid.

6. 3H4SiO4 - 4H2O = H4Si3O8(3SiO2,2H2O) : metatrisilicic acid.
Structural formulae of these acids may easily be derived, but are

purely speculative :

OH, /OH
1. H4Si04 >Si< . 2. H2Si03 SiO(OH)2.

OH/ NDH
3. H6Si2O7 (OH)3Si-O-Si(OH)3. 4. H2Si2O6 OH-O-Si-O-Si'O-OH.

5. H8Si3O10 (OH)3Si-O-Si(OH)2-O-Si(OH)3.

/O. ,0.

6. H4Si308 (OH)2Si < >Si< >Si(OH)2.

\n/ \n/

Esters of ortho- and meta-silicic acids, with known molecular
weights, have been prepared : Si(OEt)4, SiO(OEt)2.

Examples of silicates occurring in rocks, belonging to the six
classes, are given below :

1. Orthosilicates : zircon, ZrSiO4 ; olivine, Mg2SiO4 ; garnet,

Ca3Al2(SiO4)3 ; willemite, Zn2SiO4 ; potash mica, KH2Al3(SiO4)3.

2. Metasilicates : wollastonite, CaSiO3 ; leucite, KAl(SiO3)2 ; beryl,

Be3Al2(SiO3)6 ; enstatite, MgSiO3 ; talc, H2Mg3(SiO3)4 ; asbestos,
Mg3Ca(Si03)4.

3. Orthodisilicates : barysilite, Pb3Si2O7 ; serpentine, Mg3Si2O7 +

2H2O ; kaolinite, Al2Si2O7 H- 2H2O.

4. Metadisilicates : millerite, Al2HKCa2(Si2O5)6 ; petalite, LiAl(Si2O5)2.

5. Orthotrisilicate : melilith, Ca4Si3O10.

6. Metatrisilicates : orthoclase, KAlSi3O8 albite, NaAJSi3O8.

Silicates not comprised in these six groups are usually considered as
basic salts : e.g., cyanite, (AlO)2SiO3 ; andalusite, Al(AlO)SiO4.

EXAMPLE. — Calculate the formula of the mineral silicate of the
following composition :

SiO2 45-07 -j- 60-4 = 0-746 6
A12O3 38-41 -r 102-3 = 0-375 3
K2O 12-10 ~ 94-3 = 0-128 1
H2O 4-42 -4-18 = 0-245 2

The formula is therefore 6SiO2,3Al2O3,K2O,2H2O, or Si3O12Al3KH2,
or Al3KH2(SiO4)3, an orthosilicate.

xxxv BORON AND SILICON 747

Silicon. — Silicon has a great affinity for oxygen, so that the direct
reduction of silica can be effected only by the use of powerful re-
ducing agents, or at high temperatures. Silica is reduced when heated
with carbon in the electric furnace, and silicon is manufactured
in this way at Niagara by heating a mixture of sand and crushed
coke in the proportions for the reaction : Si02 + 2C == 2CO + Si,
or by reducing silica with calcium carbide. It is obtained as a hard
grey crystalline mass, with the appearance and electric conduc-
tivity of graphite, and is used in the preparation of alloys (silicon-
bronze ; manganese-silicon-bronze), on which it confers the pro-
perties of hardness and tensile strength. Silica is also reduced
when heated with carbon and iron in the blast furnace, and cast
iron, therefore, always contains silicon. Iron containing carbon and
more than 15 per cent, of silicon is very resistant to the action of
acids.

In the laboratory, silicon is most conveniently prepared by heating
silica with magnesium powder : Si02 -f- 2Mg = 2MgO -f- Si.

EXPT. 312. — A mixture of powdered quartz, or thoroughly dried
amorphous silica, with the requisite amount of magnesium powder and
one-fourth the weight of calcined magnesia to moderate the reaction,
is carefully heated in a covered porcelain crucible. The mass glows
when reaction occurs. After cooling, the magnesia is dissolved out by
hydrochloric acid, and the silicon washed in a platinum dish with
hydrofluoric and sulphuric acids to remove silica. It has a purity of
96-97 per cent. If washed on a filter, it begins to form a colloidal
solution, as in the case of boron (p. 735).

Amorphous silicon, prepared by the above process, is a dark
brown hygroscopic powder, sp. gr. 2-35, which burns brilliantly
when heated to dull redness in oxygen. When heated in air, it
burns superficially. It ignites spontaneously in fluorine, forming
the fluoride, SiF4, and burns when heated in chlorine, with pro-
duction of the tetrachloride, SiCl4. Amorphous silicon is insoluble
in water and all acids except hydrofluoric ; it is slowly attacked
by steam at a red heat : Si + 2H20 — Si02 + 2H2. A mixture
of potassium chlorate and nitric acid has no action upon it (cf.
carbon), but it dissolves readily in concentrated caustic alkalies
(cf. p. 183), or when fused with sodium carbonate, potassium nitrate,
or potassium chlorate : Si + 2KOH -f- H20 = K2Si03 + 2H2.

When amorphous silicon is strongly heated in a closed crucible,
it fuses, and on cooling solidifies to the dense crystalline graphitoidal
silicon, which also results from the reduction of silica in the electric
furnace. Octahedral crystals of silicon, orange or black in colour, are
produced by strongly heating potassium silicofluoride, K2SiF6, with
zinc or aluminium in an iron crucible, and treating the mass with
acid : 3K2SiF6 + 4A1 = 4A1F3 -f 3Si + 6KF. Zinc gives long needle-

748 INORGANIC CHEMISTRY CHAP.

shaped crystals (adamantine silicon) ; aluminium, six-sided plates
(graphitoidal silicon) ; both varieties are made up of regular octa-
hedra. Crystalline silicon has a density of 249 ; it does not burn
in oxygen, even when strongly heated, but burns in chlorine, and
ignites in fluorine. When very strongly heated, it forms grey
nodules of sp. gr. 3-0. It is attacked by a mixture of nitric and
hydrofluoric acids, or by fusion with alkali : Si + 2NaOH -f- H20 =
Na2Si03 -j- 2H2. When fused with sodium carbonate, it displaces
carbon : Si + Na2C03 = Na2SiO3 + C. Another variety (sp. gr.
2 -42) appears to be formed on crystallising from molten silver.
It is insoluble in hydrofluoric acid.

Silicon hydrides. — Silicon and hydrogen combine partly at the
temperature of the electric arc, forming silicon hydride, SiH4,
silico-methane, or monosilane : Si + 2H2 ±=; SiH4. If magnesium
powder and amorphous silica, in the proportions of 2 : 1 by weight,
are heated in a crucible, magnesium silicide, which probably consists
mainly of Mg2Si, is formed as a blue crystalline mass. This, when
treated with dilute hydrochloric acid in a flask from which air has
been displaced by hydrogen, evolves a gaseous mixture of silicon
hydrides with hydrogen, which is spontaneously inflammable :
Mg2Si + 4HC1 = 2MgCl2 + SiH4 (Buff and Wohler, 1857). If
the gas is bubbled through water, each bubble ignites in contact
with the air, and burns with a luminous flame, producing a vortex
ring of finely-divided silica : SiH4 + 202 = Si02 + 2H20 (cf.
phosphoretted -hydrogen).

If the gas, after washing with water and drying with calcium
chloride and phosphorus pentoxide, is passed through a tube cooled
in liquid air, a mixture of hydrides of silicon is condensed, and from
the liquid, by fractionation, the following compounds may be
isolated :

1. Monosilane, SiH4, m.-pt. — 185°, b.-pt. — 112°, a colourless gas,
stable at the ordinary temperature, spontaneously inflammable if
mixed with the other hydrides, and sometimes if pure. The relative
density is 16-02. It is decomposed when passed through a red-hot
tube, yielding twice its volume of hydrogen : SiH4 = Si + 2H2. By
the action of caustic alkalies, four times the volume of hydrogen is
produced : SiH4 + 2KOH + H2O = K2SiO3 + 4H2. The gas pre-
cipitates copper silicide, Cu2Si, from copper salts, and silver from silver
salts : 4AgNO3 + SiH4 = Si + 4Ag + 4HNO3.

Pure monosilane is obtained by heating triethyl silico-Jormate with
sodium : 4SiH(OC2H5)3 = SiH4 + 3Si(OC2H5)4 (ethyl orthosilicate).
The triethyl silico -formate (which is the silicon analogue of orthoformic
ester, CH(OC2H5)3) is obtained by the action of silicon chloroform
on absolute alcohol, or sodium ethoxide, NaOC5H5 :

SiHCl3 + 3C2H6-OH = SiH(OC2H6)3 + 3HC1.

xxxv BORON AND SILICON 749

2. Disilane, Si2H6 (silicon- ethane), which is also formed by the
action of concentrated hydrochloric acid on lithium silicide : Li6Si2 -}-
6HC1 = GLiCl + Si2H6, is a colourless gas,b.-pt. — 15°, m.-pt. — 132-5°,
which is stable at the ordinary temperature, but rapidly decomposes at
300°. Its relative density is 31-7. Disilane inflames in the air, is
soluble in benzene and carbon disulphide, and is decomposed by
alkalies : Si2HG + 2H2O + 4KOH = 2K2SiO3 + 7H2.

3. Trisilane, Si3H8, is a colourless liquid, b.-pt. 53°, m.-pt. — 117°,
decomposing spontaneously at the ordinary temperature. Si3H8 and
Si2H6 react vigorously with carbon tetrachloride and chloroform :
2CC14 + Si2H6 = 2SiCl4 + 20 + 3H2.

4. Tetrasilane, Si4H10, b.-pt. 80-90°, m.-pt. — 93-5°, is less stable
than Si3H8.

5. Solid hydrides, probably Si5H12 and Si6H14, remain after frac-
tionation.

The existence of silicon- acetylene, Si2H2, said to be formed as a yellow
solid by the action of hydrochloric acid on calcium silicide, is doubtful.
It has been stated to be H3Si3O2, silicone, which on exposure to sunlight
gives off hydrogen and leaves black Si3O2.

By the action of silane on solid bromine at — 80°, the substitution
products SiH3Br (m.-pt. — 94°, b.-pt. 1-9°) and SiH2Br2 (m.-pt. — 70-1°,
b.-pt. 66°) are formed. By the action of water on SiH3Br a colourless,
odourless, combustible gas, disiloxane, (SiH3)2O, rn.-pt. — 144°, b.-pt.
— 15-2°, is produced.

Halogen compounds of silicon. — Compounds of silicon, of the
types SiX 4 and SiHX3, with all the halogens are known ; isolated
compounds of the types SiH2X2 and SiH3X have been prepared. A
number of chlorides not corresponding with the type SiX4 are also
known, e.g., Si2Cl6, Si3Cl8, Si4Cl10, Si5Cll2, Si6Cl14.

Silicon tetrachloride, SiCl4. — This compound (Berzelius, 1823) is
produced when amorphous silicon, or the mixture of this with
magnesia obtained by heating 40 gm. of dry powdered sand with
10 gm. of magnesium powder, is heated in a current of dry chlorine :
Si + 2C12 = SiCl4. Chlorine may also be passed over heated
silicon-iron. An older method of preparation is to heat an intimate
mixture of silica and carbon to whiteness in a porcelain tube in a
stream of chlorine : SiCl4 <- !2C12" +"Si!O7 + "2C| -> 2CO. The pro-
ducts of reaction are cooled in a worm-tube, when silicon tetra-
chloride condenses as a colourless volatile liquid, sp. gr. 1-524,
m.-pt. — 89°, b.-pt. 56-9°, which fumes strongly in moist air owing
to hydrolysis : SiCl4 + 4H2O = H4SiO4 + 4HC1. When the gas
is passed into water, gelatinous silica is deposited. Silicon tetra-
chloride combines with gaseous ammonia, forming a white amor-
phous solid, SiCl4,6NH3.

750 INORGANIC CHEMISTRY CHAP.

By the action of chlorine on silicon, besides SiCl4, two other
chlorides are formed : the trichloride, Si2Cl6 (b.-pt. 147°), and the
octachloride (b.-pt. 210-215°). These may be separated by frac-
tionation. The trichloride, Si2Cl6, is also produced when the
vapour of the tetrachloride is passed over strongly -heated silicon.
It is a colourless, fuming liquid, b.-pt. 147°, m.-pt. — 1°, the hot
vapour of which ignites spontaneously in the air. With water, it
produces an explosive white solid, Si2H904, or (SiO-OH)2, silicon-
oxalic acid : Si2Clfi + 4H20 = (Si02H)2 + 6HC1. The octachloride,
Si3Cl8, forms with water a white powder, H4Si306, silicon-meso-
oxalic acid, the structural formula of which has been given as
0:Si(OH)— Si(OH)2— (OH)Si:O.

According to Troost and Hautefeuille, Si2Cl6 vapour begins to decom-
pose at 350°, and is completely dissociated at 800° : 2Si2Cl6 ^
3SiCl4 + Si. At high temperatures (1000°) reaction begins in the
reverse direction, and the vapour is stable. At lower temperatures,
apparently, silicon does not react appreciably with SiCl4.

The bromides SiBr4 (b.-pt. 153°) and Si2Br6 (solid) are formed in
the same way as SiCl4, and by the action of bromine on Si2I6, respec-
tively.

The tetraiodide SiI4 is formed from iodine vapour and silicon.
When heated with finely-divided silver at 280°, it forms the tri-iodide
by a reaction of condensation : 2SiI4 -\- 2Ag = 2AgI + Si2I6. The
tri-iodide forms splendid crystals, fuming in moist air. Si3Br8 and
Si4Br10 are formed by the action of the silent discharge on silicon -
bromoform, SiHBr3.

Six oxychlorides of silicon are said to exist. Si2OCl6 (b.-pt. 137°)
is formed on passing SiCl4 vapour over wh'ite-hot felspar. If the vapour
of this, mixed with oxygen, is passed through a heated glass tube, the
compounds Si4O4Cl8 (b.-pt. 200°), Si4O3Cl10 (b.-pt. 153°), Si8O10Cl12
(b.-pt. about 300°), Si2O3Cl2 (?, b.-pt. above 400°), and Si4OrCl2 (solid
at 400°) are stated to be formed, separable by fractionation.

Silicon chloroform, SiHCl3, b.-pt. 33°, m.-pt. — 134°, sp. gr.
1-3438 (cf. chloroform, CHC13), discovered by Buff and Wohler, is
prepared by passing hydrogen chloride over silicon (or the mixture
of silicon and magnesia, p. 747) at a dull red heat : Si -f 3HC1 =
SiHCl3 -{- H2. The liquid condensed in a freezing mixture is
fractionated to separate the silicon tetrachloride (b!-pt. 56-9°) also
produced. Silicon chloroform is a colourless, mobile, fuming liquid,
which is very inflammable and burns with a green-edged flame,
emitting white fumes of silica. A .mixture of the vapour with air
or oxygen explodes when brought in contact with a flame. At 800°
the vapour decomposes : 4SiHCl3 ^± 3SiCl4 + Si + 2H2.

By the action of ice-cold water on silicon chloroform, orthosilico-
formic acid, or leucone, SiH(OH)3, is formed, which readily loses water
by two molecules condensing to give silicoformic anhydride, H2Si2O3.

xxxv BORON AND SILICON 761

This is a powerful reducing agent (cf. formic acid) : H2Si2O3 + O2 =
2SiO2 + H2O (cf. H-CO2H + O = CO2 + H2O). It is readily decom-
posed by dilute alkalies, with evolution of hydrogen : H2Si2O3 + H2O —
2SiO2 + 2H2. On heating, silicoformic anhydride decomposes ulti-
mately into silica, silicon, and hydrogen : 2H2Si2O3 = SiH4 + 3SiO2 =
Si + 2H2 + 3SiO2.

Silicon bromoform, SiHBr3 (b.-pt. 116°, m.-pt. < — 60°), is formed
by the action of hydrogen bromide on silicon ; silicon iodoform, SiHI3
(b.-pt. c. 220°), is formed by the action of a mixture of hydrogen iodide
and iodine on silicon. Numerous mixed halogen compounds of silicon,
e.g., SiCl3Br, are known.

Silicon fluoride, SiF4. — The amorphous and crystalline varieties
of silicon ignite spontaneously in fluorine, forming gaseous silicon
fluoride, SiF4. Pure silicon fluoride is obtained by heating barium
fluosilicate : BaSiF6 = BaF2 + SiF4. The gas is more conveniently
prepared by the action of hydrofluoric acid on silica (Scheele, 1771) :
Si02 -f 4HF = SiF4 -f 2H20. Since it is decomposed by water (see
below), some dehydrating agent is added. Usually a mixture of
powdered fluorspar and white sand in equal proportions is heated in
a glass flask with three times its weight of concentrated sulphuric
acid : 2CaF2 + 2H2S04 + Si02 = 2CaS04 + SiF4 + 2H20. The
colourless gas, which fumes strongly in moist air, is collected over
mercury. To free it from hydrogen fluoride, it may be passed over
sodium fluoride. Silicates, such as glass, are also decomposed by
hydrofluoric acid, with evolution of silicon fluoride.

Silicon fluoride is a colourless, incombustible, strongly fuming
gas, with a normal density of 4-693 gm./lit. It solidifies, without
previous liquefaction, at — 97° under atmospheric pressure. The
solid melts at — 77° under 2 atm. pressure, and the liquid boils at

- 65° under 1810 mm. pressure. Ammonia solution decomposes
it with separation of gelatinous silica : SiF4 + 4NH4OH =
Si(OH)4 + 4NH4F.

When silicon fluoride is passed over heated silicon, a subfluoride
( ? Si2F6) is said to be formed as a white powder, which reduces potassium
permanganate solution.

Silicon fluoride forms with ammonia gas a white crystalline com-
pound, SiF4,2NH3.

The compound SiHl?3, silico-fluoroform, analogous to silicon chloro-
form, is obtained by the action of stannic fluoride, or titanium tetra-
fluoride, on the latter, and is a combustible gas, b.-pt. — 80-2°, m.-pt.

- 110°, which decomposes on heating : 4SiHF3 = 3SiF4 + 2H2 + Si,
and on contact with water :

2SiHF3 + 4H20 = Si(OH)4 + H2SiF6 + 2H2.

Hydrofluosilicic, or silicofluoric, acid, H2SiF6. — The reaction
between silicon fluoride and water, discovered by Scheele in 1771,

752 INORGANIC CHEMISTRY CHAP.

but only completely explained by Berzelius in 1823, leads to the
formation of gelatinous silica and a new soluble acid, H2SiF6
(or SiF4-2HF), called hydrofluosilicic acid, or silicofluoric acid :
3SiF4 + 4H20 = Si(OH)4 + 2H2SiF6. If the gelatinous liquid so
formed is treated with hydrofluoric acid until the silica is just dis-
solved, more hydrofluosilicic acid is formed, and the difficult process
of filtration is avoided : Si(OH)4 -f 6HF = H2SiFc -f 4H2O.

EXPT. 313. — Heat a mixture of 50 gm. of powdered fluorspar, 50 gm.
of fine white sand, and 100 c.c. of concentrated sulphuric acid in a stout
glass flask (thin glass is soon perforated) on a sand-bath, and pass the
silicon fluoride (fuming in air) into water in a cylinder, the gas delivery

tube dipping under
an inch -of mercury
at the bottom to
prevent the tube
becoming choked
by the gelatinous
silica (Fig. 369).
The latter is de-
posited in strings
of small sacs, each
enclosing a bubble

of gas ; these should
be broken down
occasionally by
stirring with a
glass rod. The
liquid is then fil-
tered through

linen, and the
silica, when

washed, dried, and
heated, is very pure
(sp. gr. 2-2).

Priestley, in describing this experiment ("Observations on Air"),
remarks : "I have met with few persons who are soon weary of looking
at it, and some could sit by it almost a whole hour and be agreeably
amused all the time."

A concentrated solution of hydrofluosilicic acid fumes in the air.
If silicon fluoride is passed into concentrated hydrofluoric acid
cooled in ice, crystals of H2SiF6,2H2O, m.-pt. 19°, separate. When
solutions of the acid are titrated with alkali the following reactions
occur :

FIG. 369.— Preparation of Hydrofluosilicic Acid.

1. H2SiFfl

2. Na2SiF6

2NaOH = Na2SiF6 (pp.)
4NaOH == 6NaF Si

2H20.
i(OH)4 (pp.).

xxxv BORON AND SILICON 753

The end-point is therefore reached, with phenolphthalein, when
six molecules of base have been added per molecule of acid.

Pure hydrofluosilicic acid does not corrode glass, but on evapora-
tion it decomposes : H2SiF6 ^± SiF4 + 2HF, and the hydrofluoric
acid set free corrodes a flask or porcelain basin. With steam at
high temperatures, crystals of silica are formed.

Hydrofluosilicic acid is obtained as a by-product in the manu-
facture of superphosphate by treating minerals containing apatite
with sulphuric acid (p. 849).

Salts of hydrofluoric acid are called silicofluorides, or fluosilicates ;
they are prepared by adding the requisite amount of base to the
acid, or by the action of gaseous silicon fluoride on the solid fluorides :
SiF4 -f- 2NaF = Na2SiF6. The following salts are difficultly soluble,
and are precipitated when hydrofluosilicic acid is added to solutions
of salts of the metals : Li2SiF6, K2SiF6, Na2SiF6, BaSiF6, CaSiF6,
YSiF6. The salts K2SiF6 and Na2SiF6 (which may be used in the
preparation of silicon by heating them with alkali-metals :
K2SiF6 -J- 4K = 6KF + Si) are formed as nearly transparent
gelatinous precipitates ; BaSiF6 forms a white crystalline pre-
cipitate ; strontium salts are not precipitated.

Silicon carbide, or carborundum, SiC. — If a mixture of sand and
crushed coke in the proportions 5 : 3, with a little salt and sawdust,
is heated electrically to 1550—2200° by a carbon rod passing through
the mass (cf. graphite), carborundum, or silicon carbide, SiC, is formed :
SiO2 + 3C = SiC -f- 2CO. This compound, discovered by Acheson
in 1891, is manufactured in large quantities at Niagara for use as
an abrasive instead of emery, since it is nearly as hard as the
diamond. The technical product is a black, coarsely-crystallised
mass exhibiting a play of iridescent colours. It is very difficultly
fusible, and may be used in furnace-linings. Carborundum resists
all reagents except fused caustic soda exposed to air, which slowly
acts upon it : SiC + 4NaOH + 202 = Na2C03 + Na2SiO3 -f 2H20.
Pure carborundum forms transparent, colourless or green, six-sided
plates, sp. gr. 3-1, and is obtained by fusing silicon with carbon
in the electric furnace.

The carborundum in the electric furnace is found to be surrounded
by a layer of siloxicon, which is said to be a definite compound, Si2OC2,
mixed with a little silicon monoxide, SiO, but may be a solid solution
of silica in silicon carbide. It is used as a refractory. A fibrous
variety, called fibrox, is used as a heat insulator instead of asbestos.

Silicon borides, SiB3 and SiB6, which are very hard, are formed in
the electric furnace. Silicon nitrides, SiN2, Si2N3, and Si3N4, are pro-
duced when nitrogen is passed over heated silicon. Silicon disulphide,
SiS2, is formed in white silky needles by heating silicon in sulphur
vapour ; it is instantly decomposed by water into sulphuretted hydrogen

3 o

754 INORGANIC CHEMISTRY CH. xxxv

and gelatinous silica. It is also formed by passing the vapour of carbon
disulphide over a strongly -heated mixture of silica and carbon :
SiO2 + CS2 + C = SiS2 + 2CO.

Organic compounds of silicon. — A few compounds have been prepared
by Kipping which contain chains or rings of alternate silicon and oxygen
atoms, and are similar to organic carbon compounds. The maximum
number of silicon atoms yet obtained in such .compounds, however, is
4, whereas hydrocarbons containing 60 carbon atoms in the chain are
known. Examples of Kipping's compounds are :

OH— Si O Si— OH C6H6X /O— Si(C0H6U

1. 2. >Si< "No.

C6H5C6H5 C6H6C6H5 C6H/ \O— Si(C6H6)/

EXERCISES ON CHAPTER XXXV

1. How are boric acid and borax obtained ? Starting with borax,
how would you prepare : (a) boric acid, (6) boron chloride, (c) boron
hydride ?

2. Describe briefly the properties of boric acid. What happens
when a solution of borax is added to : (a) concentrated hydrochloric
acid, (6) a solution of calcium chloride ?

3. How is boric acid recognised in analysis ? How and why is it
removed from solutions containing it which are to be tested for metals ?

4. Describe briefly the preparation and properties of the hydrides of
boron.

5. How is boron prepared ? What are its properties ?

6. How are boron chloride and boron fluoride prepared ? What is
the action of water on these substances ?

7. Borax, on heating, loses 47-13 per cent, of its weight of water of
crystallisation. Assuming the formula of the salt to be Na2B4O7,10H2O,
and the atomic weight of sodium and oxygen to be 22-83 and 15-88,
respectively, find the atomic weight of boron.

8. How may pure silica be obtained from a mineral silicate ? From
silica, how would you prepare : (a) silicon, (6) silicon chloride, (c) hydro-
fluosilicic acid ? Describe the properties of these substances.

9. Describe briefly the halogen compounds of boron and silicon.
In what respects do these two elements resemble, and differ from,
carbon ?

10. Describe the technical preparation of silicon and silicon carbide.
For what purpose are these substances used ?

11. In what forms does silica exist ? How are the natural silicates
classified ?

12. How is colloidal silica made ? What are 'the general properties
of colloids, and in what respects do they differ from' crystalloids ?

13. Describe briefly the preparation and properties of the hydrogen
compounds of silicon. How is silicon chloroform prepared, and what
is the action of alcohol upon it ?

CHAPTER XXXVI

SPECTRUM ANALYSIS

The spectrum. — If a solid or liquid is heated to a sufficiently
high temperature it becomes luminous. At very high temperatures,
the light emitted is white (e.g., the limelight, p. 189). Such white
light, or sunlight., when passed through a glass prism, is broken up
into a series of coloured rays, called a spectrum. In passing through
the prism the white light is sorted out into rays of different colours,
which are bent or refracted by the prism to different extents. The
red rays are the least refracted, whilst the violet rays suffer the
largest deviation. The resulting spectrum, which may be received
on a white screen (Fig. 370), shows the colours in the following
order, beginning with
the least refrangible :
red, orange, yellow,
green, blue, indigo,
and violet. This is
known as a continuous
spectrum, since the
colours shade into one
another without any
gaps. At the red end
of the spectrum, but
beyond the visible
part, there are also
rays which may be
detected by their heat-
ing effect on a thermometer with a blackened bulb. These are the
infra-red rays. Beyond the violet there are also invisible rays,
which may be detected by causing the fluorescence (p. 8) of quinine
salts and some other substances. These are the ultra-violet rays.

Each coloured ray and each kind of radiation beyond the visible
spectrum at both ends is characterised by a definite wave-length.
Light and allied invisible radiations consist of transverse vibrations
in the hypothetical ether, and the waves resulting from the periodic
vibrations differ in length according to the quality of the radiation.
The infra-red waves are the longest and the ultra-violet waves the

755 3 C 2

370. — The Spectrum .

1014to4 X 107

Blue

4*550 to 4920

X 10» to 7230
6470 to 7230

Indigo ...
Violet

4240 to 4550
3970 to 4240

5850 to 6470

Ultra-violet

600 to 3970

5750 to 5850
4920 to 5750

X- and y-rays

8-4 to 0-07

756 INORGANIC CHEMISTRY CHAP.

shortest in the spectrum. The average wave-length in the visible
spectrum is about 5 x 10~5 cm. Wireless waves are very long
waves in the ether ; X-rays and the y-rays from radium are very
short waves. Wave-lengths of radiation are usually measured in
tenth metres, i.e., 10~10 m., or Angstrom units (A.U.), as they are
sometimes called. The /A and /A/A units (p. 8) may also be used.

The following table gives the wave-lengths of all parts of the
spectrum so far investigated. The numbers range from 0-07 to
1014 A.U. ; the visible spectrum extends only over the very re-
stricted range of 4000 to 7000 A.U.

WAVE-LENGTHS IN ANGSTROM UNITS.
Wireless waves
Infra-red 3-1

Bed

Orange...
Yellow
Green ...

The gap between the ultra-violet and X-rays has been partially
bridged by short radiations recently measured in the hydrogen spec-
trum (Lyman).

Varieties of spectra. — If the light from a piece of platinum wire
heated by an electric current is passed through a prism, it is found
that at lower temperatures the red end of the spectrum alone
appears, corresponding with the red light emitted by the body.
With increasing temperature the visible spectrum extends gradually
towards the violet, and when a dazzling white light is emitted, a
continuous spectrum is obtained.

If small quantities of various salts, such as sodium, potassium,
lithium, thallium, and strontium chlorides, are heated on platinum
wires in a non-luminous Bunsen flame, it will be seen that the
different salts impart characteristic colours to the flame :

sodium salts : yellow thallium salts : green

potassium salts : purple strontium chloride : red

lithium salts : crimson calcium chloride : orange -red

If the light emitted by each of these coloured flames is passed
through a prism, the spectra produced are not continuous, but con-
sist of one or more bright lines, each corresponding with a definite
wave-length. A spectrum of this kind is known as a line spectrum,
and incandescent gases and vapours, produced by the volatilisation
of salts in the flame, differ from solids or liquids in emitting line
spectra instead of continuous spectra. No two lines given by
different elements occupy exactly the same position in the spectrum,
although they may be very close together, so that the spectrum of

xxxvi SPECTRUM ANALYSIS 75?

every element is characteristic, and may serve for the identifica-
tion of the element. This is the principle of spectrum analysis,
introduced into chemistry by Bunsen and Kirchhoff in 1860. A
chart of spectra, with the wave-lengths of the principal lines,
will be found on the inside of the front cover.

The visible spectra of salts usually correspond with those of
the metals contained in them. The spectrum of sodium chloride,
for example, is identical with the spectrum of metallic sodium.
This shows that the vapours of the salts at the high temperature
of the flame are dissociated, or broken down, into their elements.
In some cases a compound exhibits a characteristic spectrum,
superposed on that of the metal. This is the case with calcium

FIG. 371.— Band Spectra.

chloride, which first gives a spectrum of the chloride, and later a
spectrum corresponding with calcium oxide.

The spectra of compounds differ from those of elements in appear-
ance. Instead of sharp lines, the spectrum consists of broad
luminous bands, with a fluted appearance (Fig. 371), sharply defined
at one edge, called the head of the band, and shading off at the other
edge. A spectroscope of high resolving power, i.e., one which
separates the different lines as widely as possible, shows that the
bands consist of large numbers of fine lines, very close together at
the head of the band, but more and more widely separated towards
the blurred edge of the band. Fig. 371 shows the band spectrum
of calcium chloride, with the line spectrum of calcium below.

The spectroscope. — A convenient instrument for examining
spectra is the spectroscope, invented by Bunsen and Kirchhoff. The
most useful form for chemical purposes, which is that originally
used by these investigators, is shown in Fig. 372. It consists of

758

INORGANIC CHEMISTRY

CHAP,

a prism, a, of flint glass, supported on an iron stand, and a brass
tube, b, called a collimator, which is fitted at the end furthest from

the prism with an
adjustable slit, d,
shown in Fig. 373.
In this way a
narrow line of
light from the
Bunsen flame, e,
in which the sub-
stance is heated,
is focussed on the
prism, the rays
being made par-
allel by a lens in
the collimator.
The light passing
through the prism
is received by the

telescope, /, which may be moved round so as to embrace any
part of the spectrum /and contains a lens which gives a magnified
view of the spectrum in the eye-piece. In order to fix the position
of any particular line, the image of a glass scale, fixed in the third
tube, g, and illuminated by a candle or
luminous gas flame, is thrown by reflection
from the face of the prism into the telescope,
and appears above the spectrum. The
position of the line is then read off by com-
parison with this scale, and may be compared
with the positions of lines given by standard
elements.

FIG. 372. — Simple Spectroscope.

FIG. 373.— Adjustable Slit
of Spectroscope.

A convenient form of spectroscope for qualitative work is the direct
vision instrument, shown in section in Fig. 374. In this the spectrum
produced by the flint glass prisms, F, is kept in a horizontal direction by

B •-.'

FIG. 374. — Direct-vision Spectroscope.

the prisms of crown glass, C, so that a virtual image of the slit is seen by
the eye at the lens, E. This instrument is very small and handy, and
can be carried in the pocket.

XXXVI

SPECTRUM ANALYSIS

759

Production of spectra. — The spectra of gases may be observed in
the light emitted by the gas at low pressure (1-2 mm.) when sub-
jected to the electrical discharge from a coil in a Geissler tube
(Fig. 105). Volatile salts may be heated on platinum wire, moistened
with hydrochloric acid, in a Bunsen flame ; or a small fused bead
of the salt (usually the chloride) heated on the wire. The spectra
of liquids may be obtained by taking electric sparks near the surface
between platinum wires, as shown in Fig. 375, one or two Ley den
jars being put in parallel with the coil ; whilst the spectra of diffi-
cultly volatile substances are obtained by heating a small quantity
of the material in a little hollow in the lower carbon rod of the electric
arc. The spectra of some metals (e.g., iron) may be obtained by
striking an arc, or passing powerful sparks,
between rods of the substance.

If the invisible parts of. the spectrum are to be
examined, the prisms and lenses must be of rock-
salt for the infra-red, or quartz for the ultra-violet,
since these rays are absorbed by glass. The
infra-red spectrum is examined by means of its
heating effect when the radiation is absorbed by
a blackened strip of platinum called a bolometer,
the electrical resistance of which increases with the
temperature. A similar but shielded strip is
placed in the opposite arm of a Wheatstone bridge.
Langley's bolometer, used in mapping the solar
infra-red spectrum, indicated a rise of temperature
of 10~8 degrees. The ultra-violet spectrum is
rendered visible by a fluorescent screen covered
with barium platinocyanide, but is most con-
veniently recorded by its action on a photographic
plate. In this case, a camera is attached to the
spectroscope, the latter being equipped with quartz prisms. Since
the extreme ultra-violet rays are absorbed by air, or the gelatin
of a photographic plate, this portion of the spectrum ('•' Schumann
rays ") must be investigated with the whole apparatus in an
evacuated chamber, and a silver bromide film prepared without
gelatin.

Variation of spectra. — Bunsen and Kirchhoff considered that the
spectrum of an element was always exactly the same, each line
having an invariable wave-length. Pliicker and Hittorf in 1865,
however, found that nitrogen in a vacuum tube could emit two
different spectra, one a line spectrum, and the other a band spec-
trum. Both spectra may be emitted simultaneously, and the
phenomenon has been observed with many other substances.
Phosphorus emits eight different kinds of spectra. Variations of
pressure in gases lead to broadening and even to slight displace-

Fia. 375.— Appar-
atus for Produc-
ing S pa rk
Spectra.

760 INORGANIC CHEMISTRY CHAP.

ments of spectrum lines, and the invariable position of spectrum
lines under all conditions is no longer recognised. It has been found,
for instance, that slight differences exist in the positions of lines in
the iron spectrum given by the sun (see below) and by the iron-arc,
respectively. The admixture of small quantities of gases may also
appreciably alter the relative intensities (not the positions) of the
lines in the spectrum of another gas.

In the case of certain elements, the spectroscope is capable of
revealing the presence of very minute quantities of the substance —
far below the possibility of detection by chemical analysis. A
quantity of -gj^ ooo- nigm. of sodium may be detected, and all
materials show the spectrum of this element. In other cases the
spectroscope may be much less sensitive, and sometimes the spec-
trum of one substance may be practically extinguished by traces
of other substances.

The solar spectrum. — In 1802 Wollaston, examining sunlight
passing through a slit by means of a prism placed before the eye,
noticed that the spectrum was crossed by a large number of fine
black lines. These dark lines in the solar spectrum were carefully
mapped by Fraunhofer in 1814, who found that they always occurred
in the same position in the spectrum. The lines are called Fraun-
hofer's lines, and the most important are designated by alphabetical
letters. Fraunhofer suggested that they were caused by the
absorption of the particular parts of the spectrum by the passage of
the light through the atmosphere of incandescent gases surrounding
the sun. The explanation of the cause of the dark lines was, however,
first clearly stated by Kirchhoff in 1860. He brought near the slit
of the spectroscope, through which he was examining the solar
spectrum, a flame charged with sodium vapour. The two very
nearly coincident dark lines in the solar spectrum, called D by
Fraunhofer, at once changed into the two bright yellow lines of
the sodium spectrum. The latter were therefore coincident with
the dark D -lines of the solar spectrum. Kirchhoff then exchanged
the sunlight for limelight, which gives a continuous spectrum
having no dark lines. On placing a sodium flame between the
source of this light and the slit of the spectroscope, the two dark
D-lines at once appeared.

Kirchhoff observed that this result is easily explained on the
supposition that the sodium flame absorbs the same kind of rays
as it emits, whilst it is perfectly transparent to other rays. If the
intensity of the light passing through the flame is greater than that
of the same kind emitted by the flame, the absorption in the latter
will cause such a weakening of intensity in that part of the spectrum
that the lines will appear dark in contrast with the rest of the
spectrum.

If we imagine a piano played in the middle of a room which is other-

XXXVI

SPECTRUM ANALYSIS

761

wise filled with wires stretched so as to emit one particular note only,
say the c' of 256 vibrations per second, then a person outside the room
would hear all the notes except this one. These particular vibrations
are taken up by the stretched wires, which are in resonance with them,
and cause the latter to vibrate. The sound emitted by the wires is,
however, too feeble to be heard among the other louder notes, which
have not suffered absorption.

If the light emitted by a burning piece of sodium is examined
by a spectroscope, the two D-lines will be seen reversed, as dark
lines, on the background of a continuous spectrum. The solid
particles of incandescent sodium oxide produced in the flame emit
a continuous spectrum, but the sodium vapour absorbs most of the
yellow rays from this light.

EXPT. 314. — Pass a stream of hydrogen through a Woulfe's bottle in
which hydrogen is produced from zinc
and dilute hydrochloric acid containing
common salt. The gas is burnt as a
large flame, coloured yellow by sodium
from the spray, at a burner A (Fig. 376).
A small Bunsen burner B, with a head of
sodium chloride, is placed in front of the
large flame. The outer edge of the small
flame appears dark against the bright
yellow background.

The presence of sodium vapour in
the atmosphere of the sun may there-
fore be inferred from the dark lines in
the spectrum. The bright parts of the
spectrum teach us nothing as to the
elements present in the sun, because
they are merely parts of the continuous
spectrum emitted by any solid body
raised to incandescence. It has been shown that a sufficiently
thick layer of incandescent gas will also emit a continuous spectrum,
and this probably corresponds with the constitution of the sun. It
is the dark lines of the spectrum, corresponding with absorption in
the solar atmosphere, which indicate the presence of corresponding
elements in the latter. By examining these lines the composition
of the sun, given on p. 32, has been discovered.

Certain stars and nebulae, however, show bright lines on a dark
ground. These correspond with elements present in the masses
of incandescent gas or vapour.

The spectroscope, therefore, opened the way to the chemical
examination of bodies in space ; the rays of light coming from the
most distant stars reveal the chemical composition of the luminous
matter with as much certainty as if the millions of miles of inter-

FlG. 376. — Reversal of Sodium
Line in Spectrum.

762

INORGANIC CHEMISTRY

CHAP.

vening space had been annihilated, and a sample of the star placed
on the bench in the laboratory.

Absorption spectra. — If white light is transmitted through a
transparent coloured body, such as ruby glass, or a solution of
indigo, the emergent light, if examined by the spectroscope, is
found to have lost certain portions of the spectrum. These con-
stituents have been absorbed by the body, and the remaining part
of the spectrum corresponds with the colour of the body. A solu-
tion of copper sulphate removes all the spectrum except the blue
end : a solution of potassium dichromate removes all except the

red end. In other cases dark
bands, corresponding with
absorption, cross the spectra
at various parts.

The absorption spectrum
differs in most cases from the
emission spectrum of the same
substance. The dark absorp-
tion lines of chlorine gas are
not even analogous to the
bright lines in the emission
spectrum. In the case of
iodine, however, the two sets of
lines correspond. The absorp-
tion spectra of solutions are
nearly always made up addi-
tively of one or two sets of
bands, corresponding with one
or both of the two ions, respec-

FIG. 377.— Absorption Spectra of Blood.

characteristic of the ion MnO/

tively. All permanganates, for
example, show the same bands,
With concentrated solutions the

absorption due to the undissociated molecules makes its appearance,
and in the case of the nitrates, each salt shows a characteristic
ultra-violet absorption spectrum, differing according to the metal
present.

The absorption spectra of blood are shown in Fig. 377. No. 1 shows
two dark bands, D and E, due to oxy haemoglobin, given by oxidised
blood. No. 2 shows the absorption spectrum of de-oxidised blood, in
which there is only one dark band, due to haemoglobin. By the action of
acids on blood, the haemoglobin is converted into haematin, the oxidised
and de-oxidised forms of which give the spectra Nos. 3 and 4. Carbon
monoxide and hydrocyanic acid form compounds with haemoglobin,
giving characteristic absorption spectra.

Determination of wave-lengths by the spectroscope. — The position

xxxvi SPECTRUM ANALYSIS 763

of any spectrum line is determined by the scale in the instrument,
the position marked 50 being adjusted on the double sodium line.
The scale-readings, however, vary with the particular type of glass
used in making the prism, i.e., with the dispersion of the prism, and
these numbers are therefore arbitrary. The real characteristic
of a spectrum line is the wave-length of the light producing the line,
and in the identification of substances it is necessary to find the
wave-lengths of the lines in its spectrum, and compare these with
the tables of wave-lengths, or with the spectrum chart on p. 1041.
The wave-length is obtained by interpolation on a wave-length
curve. The positions of the lines on the arbitrary scale are plotted
as abscissae, and the wave-lengths of standard lines, the position
of which is found also on the arbitrary scale of the spectroscope, are
plotted as ordinates. If the ordinates are joined in a smooth curve,
the ordinates of the points where verticals from the scale readings
cut the curve give the required wave-lengths. The standard lines
shown in the chart may be used. The colour of a line may be
inferred from its wave-length by means of the list on p. 756.

EXERCISES ON CHAPTER XXXVI

1. Describe the chief characteristics of the spectra of (a) an incan-
descent solid, (b) an incandescent gas. What use is made of these
in chemistry ?

2. What are absorption spectra ? What regularities have been
noticed in the absorption spectra of salts, and how are they explained ?

3. What is known of the composition of the sun, and other stars ?
How has this knowledge been obtained ?

CHAPTER XXXVII

METALS AND ALLOYS

Metals. — The metals gold, silver, copper, iron, tin, and lead were
known to the ancients : they are mentioned in the Old Testament,
and by early Greek authors. Mercury is mentioned by Aristotle
(B.C. 384-322). Zinc is referred to by Paracelsus (1539), and bismuth
by Agricola (c. 1530). Antimony and its compounds are carefully
described by the supposed Basil Valentine. The remaining metals
have all been discovered since the seventeenth century. Mercury
was definitely included among the metals only after its solidification
by cold, which was noticed in a severe Russian winter by Braune,
in 1759.

The metals occur chiefly in veins traversing granitic, or limestone,
rocks ; more rarely in detached nodules in alluvial strata. Only a
few, viz., gold, silver, copper, mercury, and the platinum metals,
occur in the metallic, or native, state ; the rest occur as ores,
mostly oxides and sulphides, or carbonates and sulphates.

The general properties of metals have been referred to (p. 450). The
first distinct definition of a metal was, apparently, given by the Latin
Geber (p. 29) : "A metal is a miscible and fusible body, which is
extensible in all directions under the hammer." This excludes the
brittle metals, which were classed as semi-metals. Fourcroy (1789)
pointed out that such distinctions are too arbitrary to be of use :
between the perfectly malleable gold and the brittle antimony there
are insensible gradations. The exact characteristics which separate
metals from non-metals cannot, in fact, be described, and an
element like tellurium may be regarded either as a metal (from its
physical properties), or as a non-metal (from its chemical analogies
to sulphur).

The alchemists regarded metals as compounds of mercury and sulphur
(p. 29), and this idea lasted until the end of the seventeenth century.
Thus, Wilson ("Compleat Course of Chymistry," London, 1721) speaks
of gold as : " by Nature generated of a most pure fixed Mercury, and a
small quantity of clean fix'd Sulphur, of most pure Redness, which
tingeth the Mercury." The sulphur and mercury of metals were not,
however, regarded as the ordinary materials, but were occult principles.

764

CH. xxxvn METALS AND ALLOYS 765

Thomas Norton (1477) says, in connection with the alchemical princi-
ples : " Our gold and silver are not those you can hold in the hand."

Stahl, and the phlogistonists, considered the metals to be compounds
of phlogiston with the calx of the metal (i.e., its oxide). Lead is con-
verted by heating in the air into a dross, or calx, and it was supposed
that phlogiston escaped. By heating the calx of lead with charcoal, a
substance rich in phlogiston, the metal was revived : metal = calx -+-
phlogiston.

Lavoisier (1787) recognised the elementary character of the metals,
and gave a list of the seventeen metals then known, in his tables of the
elements.

The existence of allotropic forms of some metals (e.g., gold) was
discovered by Matthiessen ; more recent investigations have shown that
several metals can exist in allotropic forms. Some of these are well-
defined : ordinary tin forms a grey modification on cooling, and three
kinds of iron are recognised. In other cases the existence of allotropy
is only inferred from peculiarities in the expansion of the metal by heat,
and the different forms have not been isolated.

Alloys. — If two or more metals are fused together they usually,
but not always (e.g., zinc and lead, p. 821), form a homogeneous
liquid, and the intimate association of the metals which is formed on
solidification is called an alloy. The name alloy, which was used in
this sense by Chaucer, is derived from the Latin alligare, " to bind
to." Although the preparation of alloys by fusion is the method
most commonly used, the strong compression of finely -powdered
metals, the simultaneous electro-deposition of the metals from a
mixed solution (e.g., copper and zinc, in the form of brass, from a
solution of the cyanides in potassium cyanide), and the reduction
of one or more of the metals from compounds in the presence of the
other metal (e.g., aluminium from the oxide by carbon in the electric
furnace in presence of copper to form aluminium bronze), are
alternative processes leading to the formation of alloys. Alloys
containing mercury are called amalgams, a word which may have
been derived from Arabic.

The solid formed by the solidification of a fused mixture of metals
may be either (a) homogeneous, or (6) heterogeneous.

The homogeneous solid alloy may be :

(i) a solid solution,
(ii) a pure chemical compound, or
(iii) a solid solution of a compound in excess of one of the metals.

Compounds of metals with non-metals may also be present in
alloys ; e.g., hard steel, prepared by quenching, is a solid solution of
iron carbide, Fe3C, in a particular allotropic form of iron (y-iron).

766

INORGANIC CHEMISTRY

CHAP.

Allotropic forms of some metals, which differ from the ordinary form,
may occur in the alloys.

An alloy of platinum and silver may dissolve completely in nitric
acid, whilst platinum itself is insoluble ; an alloy of 10 per cent, of gold
with potassium, when thrown into water, leaves the gold as black
powder, which forms a colloidal solution with water ; this form is con-
verted into ordinary yellow gold by heating to redness.

If the solid alloy is heterogeneous, the separate phases (p. 106) may
consist of :

(i) pure metals,

(ii) one or more pure compounds, or
(iii) solutions of metals, or compounds, in metals.

Freezing-point curves of alloys. — The class to which an alloy
belongs may be determined by an examination of the freezing points
of fused mixtures of the constituents in various proportions. For
simplicity we shall consider only two components, X and Y,
forming a binary alloy, and shall suppose that this alloy either
(a) is a heterogeneous mixture of the two pure components, or
(6) consists of one or more chemical compounds, with or without
an excess of one of the pure components. The consideration of
solid solutions is more difficult, and is omitted.

We consider first the case in which no chemical compounds of

the metals are present in the
alloy. If the pure metal, X,
say silver, is fused, and allowed
to cool, it will begin to solidify
(if supercooling is absent) at
the freezing point. This may
be represented on a diagram
(Fig. 378) by the point A. In
the diagram, the temperature
of solidification is measured
vertically, and the composition
of the alloy is represented on
the horizontal by dividing the
latter into 100 parts, each
representing one atomic pro-
portion in 100 atomic propor-
tions of total alloy. Thus, the

*100 90 80 70 60 50 40 30 20 10 0 ^ p> rf ^ jj^ mugt

correspond with 0 part of the
second metal, Y, say lead, and
the point A on the vertical line above P represents the melting
point of pure silver (atoms of lead = 0 ; atoms of silver = 100).
The point Q will then represent pure lead (atoms of lead = 100 ;

10 20 3O 40 50 60 70 80 90 100 Y

Fia. 378.— Freezing Point Curves of Binary
Alloy Forming Eutectic.

xxxvii METALS AND ALLOYS 767

atoms of silver = 0), and B is the melting point of pure lead. A
point midway on the line PQ represents a mixture of equiatomic
amounts of lead and silver (atoms of lead = 50 ; atoms of silver =
50). The proportions of lead in the alloys are represented on the
upper scale, those of silver on the lower scale ; the sum is always
100.

If a little lead is added to the silver, the fused alloy will begin to
solidify at a temperature slightly lower than the melting point of
pure silver, since a dissolved substance (in this case lead) lowers the
freezing point of a solvent (in this case silver), provided the pure
solvent separates on cooling (p. 104). This temperature will be
represented by a point a little to the right of A. By adding suc-
cessive amounts of lead, the freezing points become progressively
lower, and they will lie on a curve such as AE. If the molecular
depression of freezing point were constant, this would be a straight
line, which has been drawn in Fig. 378 for simplicity ; usually AE
is not straight, since the laws of dilute solutions do not apply
strictly. The depression of freezing point will continue as more
lead is added, until at a certain point. E, both silver and lead begin
to separate side by side. This is the eutectic point (p. 104) , and is the
lowest temperature at which solidification of the alloy can begin.
A mixture containing the metals in the proportions corresponding
with the eutectic mixture (60 : 40 in the diagram) will solidify com-
pletely at the eutectic temperature.

Exactly the same conditions apply to the addition of silver to
lead. In this case the freezing points of various alloys lie on the
curve BE, running down from the freezing point, B, of pure lead.
The eutectic point, E, is again reached, when silver begins to
separate along with the lead.

At all points above the region AEB in the diagram the alloy is
entirely liquid ; if a horizontal line, RS, is drawn through E, then
at all temperatures included in ARE pure silver separates from
fused alloys having compositions given by the lower line beneath
RE. In the region BSE pure lead separates. At the eutectic
temperature, represented by RS, lead and silver separate together.
At temperatures below RS the whole is solid.

Now consider what occurs when a fused alloy represented by the
point D is cooled. It remains liquid until the temperature has fallen
to such a point that the curve EB is intersected at D. Since the
curve EB corresponds with separation of lead, this metal will now
begin to separate in the solid state. The still liquid part will become
richer in silver (since pure lead has separated), and will freeze at a
somewhat lower temperature. Both the composition and freezing
point will now be represented by a point on the curve nearer E.
As solidification proceeds, the temperature falls, until finally the
eutectic point E is reached, when silver begins to separate along

768 INORGANIC CHEMISTRY CHAP.

with the lead, and the whole mass then solidifies at the eutectic

temperature.

If the solid alloy resulting from the above experiment is polished,

etched with a suitable reagent, and examined under the microscope

with light reflected
from the metallic sur-
face, we shall see
crystals of lead, which
separated along D'E,
embedded in a matrix
of eutectic alloy. The
latter is always com-
posed of small crystals.
The large cubes in Fig.
379 represent the first
constituent to separate
from an alloy, em-
bedded in a eutectic
matrix.

In the second place

FIG. 379. — Microscopic Appearance of Solidified Alloy. we "will consider an

alloy in which metallic

compounds are formed, say tin and magnesium, which form Mg2Sn.
A hypothetical curve is shown in Fig. 380.

The compound of X and Y, say XY2, will have a definite melting
point, represented by C. If pure X is added to the fused compound,
or to a mixture of X and Y in
the requisite proportions to
form XY2, the freezing point is
lowered along CEly since the
compound now acts as a sol-
vent for X. The solid separating
along CEl is pure XY2. Finally,
a eutectic point, Ev is reached,
at which XY2 and X separate
together. If XY2, or pure Y,
is added to pure X, the freezing
point of the latter is depressed
along AEV the solid separating
being pure X, until E1 is again
reached, when X and XY2
separate. The solid alloy ob-
tained on cooling a liquid

FIG. 380.— Freezing-point Curves of Binary
Alloy Forming One Compound.

mixture of composition C will be homogeneous XY2. An alloy
formed by the complete solidification of a liquid of a composition
enclosed within the verticals between C and El will consist of

xxxvii METALS AND ALLOYS 769

crystals of XY2 embedded in a matrix of a eutectic mixture of XY2
and X.

Exactly similar relations hold for the addition of an excess of
Y to XY2, or XY2 to Y, when a second eutectic point E2 will appear.
Between E2 and B pure Y separates ; at Ez the eutectic XY2 -f- Y
separates.

If we commence with pure X and add increasing amounts of Y
until practically pure Y is obtained, the freezing points will make
up the whole curve AE1CE2B) which has a maximum and two
eutectics. A curve of this type is characteristic of the formation
of one compound. If there are two compounds there will be two
maxima, and so on. The rounded form of the maximum indicates
that the compound is partially dissociated in the liquid state :
XY2 ^ X + 2 Y. The microscopic appearance of a pure metal, or
of an alloy which is a definite compound, is that of more or less large
crystals which are practically in contact, since there is no eutectic
matrix.

EXERCISES ON CHAPTER XXXVII

1. Give a brief account of the various opinions which have been held
as to the nature of the metals. What are regarded as characteristic
properties of metals ?

2. What are alloys ? How are they prepared, and into what groups
may they be classified ?

3. Explain how it is possible, from the form of the freezing-point
curves, to distinguish between alloys which are mechanical mixtures and
those which contain chemical compounds.

CHAPTER XXXVIII

THE METALS OF THE ALKALIES

The alkali-metals. — Under the name alkalies are included the
substances of formula ROH, potash, soda, ammonia, lithia, rubidia,
and caesia. The metals of the alkalies are therefore potassium,
sodium, lithium, rubidium, and caesium. The radical ammonium,
NH4, although it has not been isolated, behaves in its compounds
as a univalent alkali-metal. It forms an amalgam with mercury,
and thus exhibits metallic properties, so that ammonium compounds
are usually considered with those of the alkali-metals.

The properties of the alkali-metals are shown in the table below.

Lithium.

Sodium. Potassium.

Rubidium.

Caesium

Density at 0°

0-59

0-9723

0-859

1-525

1-903

Melting point

180-1°

97-6°

62-04°

38-5°

25°

Boiling point

> 1400°

877°

758°

696°

670°

Atomic weight

(H = 1)

6-89

22-82

38-79

84-77

131-76

Atomic volume

11-7

23-5

44-4

55-8

70-2

Colour of va-

pour

purple,

green

blue

green fluor-

escence

Action on
water...

Oxides

slowly
decom-
poses

Li90

rapidly de- decom- decom-

composes, poses, poses,

but does and and •

not burn burns burns .

/Na20,
\Na202

fK20,
\K204

/Rb20,Rb202,

Rb2O4

decom-
poses,
and
burns

/Cs2O,Cs2O2,
tCs203,Cs204

The gradation in properties, with increasing atomic weight, is
clearly seen. The metals of the alkalies are the most electro-
positive elements known ; they never produce acids, or complex

770

CH. xxxvm THE METALS OF THE ALKALIES 771

anions, and -displace all other metals from their salts. In the group
itself, the electropositive character increases from lithium to
caesium, the latter being the most electropositive of the metals.
The basicity of the hydroxides increases in the same manner.

The alkali-metals are univalent, forming salts of the type RX ;
although a few higher halogen compounds are known, these
are very unstable :

LiCl4I,4H2O KI3 RbBr3 RbBr2I RbI7 CsBr3 CsCl4I

KI,(?) RbClBr2 RbCl2I RbI9 CsBr6 CsI9

KI9(?) RbCl2Br RbClBrI CsI3

KIC14 RbI3 RbCl4I CsI5

The alkali-metals all combine directly with hydrogen, forming
solid, non-metallic, hydrides, RH, decomposed by water :
RH + H2O = ROH + H2.

The vapour densities of potassium and sodium have been deter-
mined approximately, and correspond with monatomic molecules :
Na and K. In solution in tin, sodium also exists as single atoms.

Acids, bases, and salts. — Although typical representatives of these
three important classes of chemical compounds have been studied
in the preceding pages, and their general properties considered,
no attempt has been made to give logical definitions of the groups.
This is, in fact, a matter of some difficulty, since the properties of
one can hardly be specified without reference to those of the other
two members.

The ancients knew only one acid, viz., common vinegar, or crude
acetic acid, produced by the oxidation of wine, which becomes
sour on exposure to air (Greek oxos, vinegar ; oxus, sour). They
knew that vinegar effervesced with natural sodium carbonate
(nitrum, Proverbs xxv, 20), and the solvent properties of acids
figure in the story of Cleopatra and the pearl. Other acids (sulphuric,
nitric, hydrochloric) were discovered by the alchemists ; Scheele
(1770-1786) isolated a number of organic acids, i.e., acids containing
carbon, hydrogen, and oxygen, of which acetic acid, C2H402, is an
example. These acids, such as citric (C6H807), tartaric (C4H606),
and malic (C4H605), impart a sour taste to unripe fruits, whilst the
acidity of sour milk is due to lactic acid (C3H603). Boyle (1663)
recognised the following as the properties of acids :

(1) They possess a sour taste.

(2) They act as solvents, but with varying power on different bodies ;
the varying strengths of acids was recognised by Tachenius in 1666.

(3) They precipitate sulphur from a solution of liver of sulphur
(polysulphides of potassium).

(4) They turn many blue vegetable colours (e.g., litmus) red, the colour
being restored by alkalies.

3D 2

772 INORGANIC CHEMISTRY CHAP.

(5) They combine with alkalies, the characteristic properties of each
substance disappearing, and a neutral salt being formed.

On the basis of these tests, Hoffmann (1723) and Black (1755) were
able to show that carbonic acid occurring in mineral waters, is a true
acid, though a weak one.

(6) Cavendish ( 1 766) showed that hydrogen is evolved by the action of
acids (except nitric) or zinc, iron, and tin.

Examples of alkaline substances, wood ashes, and natron (native
sodium carbonate) were known to the ancients. The alchemists of the
thirteenth century were acquainted with ammonium carbonate in
the form of spirit of hartshorn, prepared by the destructive distilla-
tion of horn and bones, or the putrefaction of urine (cf. p. 801).
Later on, it was found that the salt obtained by the lixiviation
of the ashes of plants growing on ^he sea littoral had the same
properties as natron, whilst seaweeds contained the same alkali
as wood ashes. The latrochemists first described the general
properties of alkalies. These properties were found to be enhanced
by boiling with milk of lime, and the names mild alkali and caustic
alkali were introduced for the alkali before, and after, this treat-
ment, respectively.

As general properties of alkalies, the following were recognised r

(1) Their solutions feel soapy when rubbed between the fingers.
(This is probably due to corrosion of the skin, since it is felt with con-
centrated sulphuric acid ; acids when diluted usually feel very harsh
when so treated.)

(2) They restore the blue colour of dyes reddened by acids (e.g., red
cabbage, litmus), and turn the extract of violets green.

(3) They neutralise acids to form salts.

(4) The " mild " varieties effervesce with acids, giving off " fixed
air " (C02).

The difference between potash, from wood ashes, and soda, from
natron or the ashes of marine plants, was established by Duhamel
in 1737. Margraaf (1757) differentiated between potash and
soda as follows :

Potash. Soda.

1. Heat on plati- Colours the flame Colours the flame
num wire in alco- violet. yellow.

hoi flame.

2. Add p 1 a t i n i c Gives a yellow crys- Gives no precipitate,
chloride to solution talline precipitate.

in hydrochloric
acid.
Scheele found that tartaric acid gives a white precipitate of cream of

xxxvm THE METALS OF THE ALKALIES 773

tartar with concentrated solutions of potassium salts, but no precipitate
with sodium salts. The latter are precipitated by a solution of
potassium pyroantimoniate (p. 936).

Black's researches on the alkalies. — The chemical nature of the
alkalies was largely elucidated by the classical researches of Joseph
Black (6. 1728-d. 1799), ("Dissertation on Magnesia," 1754).
At that time three alkalies, and a mild and caustic form of each,
were known :

(1) Mild vegetable alkali (potassium carbonate, K2CO3), obtained by
the lixiviation of plant ashes. On boiling with lime, this gave the
caustic vegetable alkali (potassium hydroxide, KOH).

(2) Mild marine alkali (sodium carbonate, Na2CO3), obtained in
Normandy and Spain by the lixiviation of ashes of plants growing on
the sea-shore (deep-sea weeds contain the vegetable alkali). With
lime this gave the caustic marine alkali (sodium hydroxide, NaOH).

(3) Mild volatile alkali (ammonium carbonate, (NH4)2CO3), obtained
by the destructive distillation of bones, from fermented urine, or from
the sal-ammoniac of Egypt. This gave a caustic volatile alkali (ammo-
nium hydroxide, NH4OH) with lime, as was recognised by Kunckel
(" Laboratorium chymicum," published in 1716, fourteen years after
his death).

According to the phlogistic theory then in vogue, limestone on
burning absorbs phlogiston, or the " principle of causticity," from
the fire, which imparts its properties to the quicklime :

Limestone -f <£ = quick (or caustic) lime.

The process of converting a mild into a caustic alkali by boiling
with quicklime was similarly regarded as transference of phlogiston :

Mild alkali -f <£ = caustic alkali.

On boiling mild alkali with quicklime, the phlogiston was trans-
ferred from the quicklime to the alkali, rendering the latter caustic,
whilst the lime was converted into limestone :

(Limestone -f- <£) -|- Mild alkali = Limestone -f- (Mild alkali -f- <£)

or or

Quicklime Caustic alkali

Black (who worked chiefly with magnesia, the mild form of which
is easily decomposed by heat) succeeded in overturning this aspect
of the theory of phlogiston. He found that when limestone is heated
there is a loss of weight, and fixed air (CO2) is disengaged. If the
residual quicklime is dissolved in water, and boiled with mild alkali,
a weight of limestone exactly equal to that taken for calculation
in the first experiment is obtained ; it had therefore been exactly

774 INORGANIC CHEMISTRY CHAP.

reproduced by taking fixed air from the mild alkali, leaving the
latter caustic :

(1) Limestone = Quicklime -J- Fixed air (experimentally proved).

(2) Caustic alkali -j- Fixed air = Mild alkali (assumed).

(3) Quicklime -f (Caustic alkali + Fixed air) =

Mild alkali
(Quicklime -f Fixed air) -f- Caustic alkali ;

Limestone.

(agreeing with assumption (2), and offering a simple explanation of
causticising

In modern notation, these reactions are represented as follows :

(1) CaC03 = CaO + CO2.

(2) 2KOH + CO2 = K2CO3 -f H2O.

(3) CaO + H2O + K2CO3 = CaCO3 + 2KOH.

The assumption made in statement (2) was proved as follows.
The same fixed air was obtained by the action of an acid on mild
alkali as by the action of an acid on limestone, and the solution of
limestone in an acid gave the original weight of limestone when
precipitated by. a mild alkali :

CaCO3 + 2HC1 = CaCl2 + CO2 + H2O.
K2CO3 + 2HC1 =r 2KC1 + CO2 + H2O.
CaCl, + K2C03 : CaC03+ 2KC1.
Statement (3) then followed as a logical consequence of (1) and (2).

Black's results were disputed by F. Meyer (1764), whose absurd
conclusions were warmly approved by Lavoisier ; these, and other
attacks, were easily repulsed by Black, and his theory was finally
accepted by the phlogistonists themselves.

Davy's isolation of the alkali-metals. — Previous to the researches
of Davy the caustic alkalies were regarded as elements, although
Lavoisier hinted that they might be oxides of unknown metals.
Humphry Davy (b. 1778 — d. 1829), whose name is chiefly
rememembered for the invention of the safety-lamp, carried out
the earliest investigations on electrochemistry. Becoming con-
vinced of the great power of decomposition exhibited by the
voltaic battery, and attracted by Lavoisier's conjecture, Davy
attempted to decompose the alkalies by electrolysis. The experi-
ment succeeded.

In 1807 he found that : "A small piece of pure potash which had
been exposed for a few seconds to the atmosphere, so as to give con-
ducting power to the surface [by attraction of moisture, and slight
deliquescence], was placed upon an insulated disc of platina, connected
with the negative side of the battery ... in a state of intense activity ;
and a platina wire, communicating with the positive side, was brought

xxxvm THE METALS OF THE ALKALIES 775

in contact with the upper surface of the alkali. . . . The potash began
to fuse at both its points of electrization. There was a violent effer-
vescence at the upper surface ; at the lower, or negative surface, there
was no liberation of elastic fluid, but small globules having a high
metallic lustre, and being precisely similar in visible characters to
quicksilver, appeared, some of which burnt with explosion and bright
flame, as soon as they were formed, and others remained, and were
merely tarnished, and finally covered with a white film which formed
on their surfaces. These globules, numerous experiments soon showed
to be the substance I was in search of, and a peculiar inflammable
principle the basis of potash."

This metal, which Davy called potassium, was found to possess
extraordinary properties :

(1) It is lighter than water (density 0-875).

(2) When thrown on water it instantly decomposes it, attracting
the oxygen ; the liberated hydrogen is ignited by the heat developed,
and burns over the rapidly-moving floating globule of metal with a
heliotrope-coloured flame. Some of the caustic potash produced dis-
solves in the water, but a small fused globule is left, which exists in
the spheroidal condition, and, on cooling down, dissolves with a sharp
crack, often being projected from the surface of the liquid.

(3) The metal rapidly oxidises in the air ; a freshly -cut piece, which
shows a bright, metallic lustre for an instant, becoming at once covered
with a blue tarnish. The metal is therefore preserved under petroleum,
which is free from oxygen.

In the same way, from caustic soda, sodium was isolated, and by
heating these metals with the alkaline earths, lime, strontia, baryta,
and magnesia, the metallic bases of the latter were prepared, and
called calcium, strontium, barium, and magnesium. Boron was
isolated by the action of potassium on fused boric acid. Sodium,
like potassium, decomposes water, but, as the heat evolution is not
so great, the liberated hydrogen does not take fire unless the sodium
is prevented from moving about by placing it on starch -jelly ; the
hydrogen then catches fire, and burns with a bright yellow flame.

Gay-Lussac and Thenard in 1808 showed that, when molten caustic
potash or soda was brought in contact with red-hot iron turnings,
the iron was oxidised, and the alkali metal distilled off. At the same
time, a considerable amount of hydrogen was evolved. The caustic
alkalies were then recognised as hydroxides, KOH and NaOH, of the
metals potassium and sodium, not, as had been supposed by Davy,
the oxides.

EXPT. 315. — The presence of hydrogen in caustic potash or soda may
be shown by heating a mixture of the powdered alkali with iron filings
in a hard glass tube. Hydrogen is evolved, and may be ignited at the
mouth of the tube.

776 INORGANIC CHEMISTRY CHAP.

Acidic and basic oxides. — Oxides which unite with water to produce
acids and bases, respectively, are called acidic and basic oxides (p. 134).

In some cases, a basic oxide, although forming salts with acids,
does not yield an appreciably alkaline solution. This results simply
from the small solubility of the oxide, because an indicator such as
litmus or phenolphthalein does not react until the hydrogen or
hydroxide ions are present in finite, although small, concentrations,
the numerical values of which can be determined for each indicator
(p. 364).

In the case of cupric oxide, for instance, which dissolves readily
in dilute sulphuric acid to form cupric sulphate, the solubility in water
is so minute that, although the dissolved portion, even in a saturated
solution, is practically completely ionised on account of the great
dilution (p. 358), yet the total concentration of hydroxide ions never
reaches the minimum value required to change the colour of the indi-
cator. The neutralisation with acid, however, follows the normal
course, since the solution and ionisation of the base proceeds, as
hydroxide ions are removed by the acid : CuO (solid) + H2O —
Cu(OH)2 (dissd.) ^± Cu" + 2OH'.

H2SO4 ^± H' + HSO4' ^ 2H* + SO4".
H' + OH ^± H20.

The minute trace of copper oxide dissolved in water is readily detected
by its catalytic acceleration of the oxidation of sulphites by atmospheric
oxygen (p. 494).

Alumina, A12O3, dissolves both in acids and in alkalies. Aluminium
hydroxide is a very weak electrolyte, which can ionise either as an
acid or as a base. Both functions are developed simultaneously,
since the ionisation in a saturated solution never produces hydrogen
and hydroxide ions in excess of the ionisation of water :

A1203 (solid) + Aq.~2Al(OH)3(dissd.) =

Such a substance, exhibiting both acidic and basic functions, which
become perceptible in the presence of strong bases and strong acids,
respectively, is called an amphoteric electrolyte. Its salts with
strong acids and strong bases are largely hydrolysed in solution
(p. 896).

The composition of salts from acids and bases was first clearly ex-
pressed by Tachenius, who says ("Hippocrates chimicus," 1666):
" Omnia salsa in duas dividuntur substantias, in alcali et acidum."
This was the basis of the dualist ic theory (p. 274), and in another form
it appears in the modern ionic theory.

xxxvin THE METALS OF THE ALKALIES 777

SODIUM, Na = 22-82.

The alkali industry. — Sodium carbonate in a very impure form
was formerly prepared by burning plants growing on the sea-shore
(Chenopodium, Salicornia, Salsola, etc.). The plant-ash was called
barilla, and was used in the manufacture of soap. When Stahl
pointed out that the base of common salt is an alkali, attempts were
made to obtain soda from this source. An early process was that of
Scheele (1773), in which salt is decomposed by boiling with litharge :
2NaCl + 4PbO + H20 = 2NaOH + PbCl2,3PbO. The same
chemist also observed that a mixture of lime and salt, when
moistened, slowly effloresced, with the formation of sodium
carbonate. The preparation of alkali from common salt, however,
was first satisfactorily effected by Nicolas Leblanc in 1787. His
process comprised the following steps :

(1) Salt is converted into sodium sulphate by heating with sulphuric
acid : 2NaCl + H2SO4 = Na2SO4 + 2HC1.

(2) The sodium sulphate, or salt-cake (p. 238), is heated to dull
redness with a mixture of limestone and powdered coal, when sodium
carbonate and calcium sulphide are produced. The reaction probably
occurs in two stages :

(a) Na2SO4 + 2C = Na2S + 2CO2.

(6) Na2S + CaC03 = Na2CO3 + CaS.

The final product is known as black-ash ; if it is broken up and
lixiviated with water, an impure solution of sodium carbonate is
obtained, whilst the sparingly soluble calcium sulphide (with excess
of coal, limestone, and impurities) remains as alkali-waste.

Leblanc established his process in a works by means of a loan from
the Duke of Orleans in 1791. Two years later the Duke was guillotined
by the friends of liberty and fraternity, and Leblanc's factory was con-
fiscated. The unfortunate inventor, who indeed escaped the fate of his
benefactor, lingered on only to die by his own hand in 1806.

After the repeal of the salt tax in England, an alkali works was
established in Lancashire, in 1823, by Muspratt, in which the Leblanc
process was used. During the nineteenth century the Leblanc process
was one of the most important British industries, the production of
sodium carbonate in the period 1879-1883 being 500,000 tons per
annum.

The Leblanc process. — In this process, sulphuric acid, made by
the chamber-process from pyrites, is heated with salt for the produc-
tion of salt-cake (p. 238), the hydrochloric acid being absorbed and
converted into chlorine, which is mostly used in the manufacture of
bleaching powder (p. 376). From the burnt pyrites, copper, and
sometimes silver and gold, are extracted.

The reduction of the salt-cake with carbon, in the presence of

778 INORGANIC CHEMISTRY CHAP.

limestone, is carried out in black-ash furnaces. At present the
product is wholly worked up as caustic soda, NaOH, and the opera-
tion of making the black-ash is carried out in revolving furnaces.

The black-ash revolving furnace, or " revolver," consists (Fig. 381)
of a cylinder, B, of iron plates lined with firebricks, 15-20 ft. long,
running on rollers by means of bands on the outside of the cylinder.
The rotation is effected by a cog-wheel passing around the cylinder,
which engages with a smaller driving cog-wheel below. The firing
is effected by producer gas, made in a generator, A, close to the
furnace, the flame passing into the revolver through a fireclay ring
called the " eye," hung between the end of the furnace and the outlet
from the gas generator. The charge consists of 2 tons of salt-cake,
2 tons of crushed limestone, and 1 ton of coal slack, and is intro-
duced in one batch. At first the revolver is turned slowly ; it is
finally speeded up to 5 or 6 revolutions per minute, and rotation is
continued until a yellow flame of carbon monoxide appears. The
pasty mass is then discharged into iron trucks through a manhole,
about If tons of black-ash being obtained. The waste heat from
the furnace is utilised bypassing the hot gases over a series of
evaporating pans.

The cooled black-ash is broken up and lixiviated with water in
Shanks's lixiviating tanks, operated on the counter-current principle.
Fresh water is added to the tank containing nearly spent ash, and
the concentrated liquors are used in leaching the freshly added black-
ash. The liquors are conveyed from tank to tank by siphon pipes.

The insoluble residue in the lixiviators, the alkali-waste, is treated
by the Chance-Glaus process (p. 476). The liquors contain sodium
carbonate, caustic soda, and impurities such as sodium sulphide and
iron salts (p. 992) ; they are worked up directly for the production
of caustic soda, this process having been introduced in Lancashire
in 1853.

A diagrammatic scheme of the Leblanc process is given below,

PYRITES SODIUM NITRATE SALT COAL LIMESTONE

(45% S) (97%) (97%) 250 parts 120 parts

63 parts 1 part 100 parts * \

^ J/ | ^ i +

Sulphuric Acid (95%) >Salt-cake ^Crude Alkali (" Black

105 parts 120 parts 170 parts Ash")

HYDROCHLORIC
ACID (sp. gr. 1-16) i

180 parts SOD A- ASH

72 parts
BURNT PYRITES or CAUSTIC RECOVERED

45 parts SODA SULPHUR

for wet copper extraction. 60 parts 20 parts

XXXVIII

THE METALS OF THE ALKALIES

779

Caustic soda, NaOH. — The Leblanc liquors (or solutions of sodium
carbonate from the ammonia-soda process) are run into causticisers

FIG. 381. — Black Ash Revolving Furnace.

(Fig. 382), iron tanks provided with mechanical agitators, and a
pipe for admission of steam. Quicklime is placed in an iron cage
dipping into the top of the liquor, the stirrer is started, and steam is
blown in. The sodium carbonate is practically completely converted
into caustic soda :

Na2C03 + Ca(OH)2^:2NaOH + CaC03.

Calcium carbonate is slightly soluble, and the dissolved part reacts
with caustic soda, converting a portion into sodium carbonate by
the reverse reaction. As the concentration of sodium carbonate in
the solution decreases,
owing to caustifica-
tion, the solubility of
calcium carbonate in-
creases, since the C03"
ions of the sodium
carbonate, which de-
press the solubility of
the calcium carbonate,
are progressively re-
moved. At the same
time, the solubility of

u u u

u u

iffil

FIG. 382. — Causticiser.

the calcium hydroxide
decreases, since the
increasing concentra-
tion of hydroxide ions, OH', of the caustic soda depresses
the solubility of the calcium hydroxide. A state of equilibrium
is reached when the solubilities of the calcium carbonate and
calcium hydroxide become equal, since then no further conversion
of the one solid phase into the other, brought about by solution of
one and the subsequent precipitation of the other solid phase, can
occur.

780

INORGANIC CHEMISTRY

The solubilities of the two solid phases are regulated by the solubility-
product equations (p. 358) :

(1) Ca(OH)2 (solid) — Ca(OH)2 (dissd.) — Ca" -f 2OH'

.'. [Ca"]! X [OH']2 = constant = Kr

(2) CaCO3 (solid) ±^ CaCO3 (dissd.) ^ Ca" + CO3"

/. [Ca"]2 x [CO3"] = constant = K2.

For equilibrium [Ca"^ = [Ca"]2, i.e., the solubilities of the two solids
are equal,

. [OH']2
[CO/]

With increasing concentration the equilibrium is shifted from the
hydroxide side of the equilibrium equation to the carbonate side, since
the concentration [CO/] is involved as the first power in the equilibrium
constant, whereas the concentration [OH'] is
involved as the square. Caustification is more
complete (99 per cent.) in dilute solutions
(normal). The carbonate solution used in
practice has a density of 1-1, when 91-92 per
cent, of caustification is obtained.

Better results are obtained with strontia or
baryta instead of lime, since the hydroxides
of strontium and barium are more, and the
carbonates less, soluble than those of calcium.
Strontia and baryta are too expensive for in-
dustrial use.

The causticised liquor is next filtered from
the lime sludge in a vacuum filter, and
concentrated, usually in vacuum evaporators.
In the latter the liquid is heated by steam
coils or jackets under reduced pressure ; the
boiling point is lowered and steam at a lower
temperature than 100° (e.g., exhaust steam)
can be employed. One type is the Eestner
evaporator (Fig. 383), consisting of a series of
tubes in an outer jacket heated by exhaust
steam. The liquor enters inside the tubes
at the bottom, under reduced pressure. It
commences to boil, and the foam is projected
into a collecting head, where it is given a
rotary motion by means of vanes. The
concentrated liquor thus separated runs off, whilst the steam
passes out, either to a similar apparatus under still lower
pressure or to a condenser, where it is condensed by cool-
ing, say with a jet of water. The air from the cooling water is

Fia. 383. — Kestner
Vacuum Evaporator.

xxxvin THE METALS OF THE ALKALIES 781

removed by a vacuum pump which maintains the low pressure

in the apparatus.

The concentrated solution begins to deposit sodium chloride,

carbonate, etc., which are removed, and the clear liquor is finally

heated in hemispherical cast-iron soda-pots over a free flame until all

the water is driven off, and fused caustic soda remains. This is

ladled out into iron barrels, where it solidifies.

In the case of Leblanc soda, a little sodium nitrate is added to the

fused charge, to oxidise sulphides and cyanides. Graphite is formed

from the latter.

For laboratory purposes the caustic soda is fused and cast into

sticks, or powdered. The latter form is usually purer, and is more

convenient for use.

In purifying commercial caustic soda (or potash) containing chloride,
carbonate, and sulphate, it is warmed with alcohol. The impurities
do not dissolve, and the solution is decanted into a silver dish, evapo-
rated, and the residue fused (Berthollet). This material is sold as
pure by alcohol. It may contain sodium nitrite, and sodium acetate,
formed from the alcohol during the evaporation. The purest
caustic soda is made from metallic sodium. A piece of sodium which
has not been kept under oil is squeezed through a sodium press into
distilled water, previously boiled and cooled, contained in a silver dish.
The sodium wire should be lowered slowly into the water, so that pieces
do not become detached. The solution is evaporated arid the residue
fused.

Caustic soda is a white, slightly translucent, solid with a fibrous
texture. It fuses at 310°, and at about 1300° it dissociates into its
elements : 2NaOH ^ 2Na + H2 -f- 02. When exposed to the air,
it first deliquesces from absorption of moisture and a little carbon
dioxide, forming a saturated solution. The latter, however, slowly
resolidifies from absorption of carbon dioxide, when the sparingly
soluble bicarbonate, NaHC03, is formed. (Caustic potash does not
resolidify, since potassium bicarbonate is readily soluble. For this
reason a concentrated solution of caustic potash is used in gas analysis
to absorb carbon dioxide, since it does not deposit solid, which would
choke the apparatus.) Caustic soda is a powerful cautery, breaking
down the proteins of the skin and flesh to a pasty mass.

Several hydrates of caustic soda, e.g., NaOH,H20, m.-pt. 64° ;
NaOH,2H2O, m.-pt. 12°, have been described.

The chief use of caustic soda in commerce is in the manufacture
of soap (p. 206). Fats are boiled with caustic soda until hydrolysis
has occurred, and the soap, which consists of sodium salts of the
fatty acids, is separated by adding salt (" salting out "), when it
rises to the surface, is removed, pressed, and cut into bars.

The ammonia-soda process.— In 1838 Dyar and Hemming pro-

782 INORGANIC CHEMISTRY CHAP.

posed to make sodium carbonate from common salt by precipitating
a concentrated solution of the latter with ammonium hydrogen
carbonate, when sodium hydrogen carbonate (" sodium bicarbonate,"
or " bicarbonate of soda ") separates out :

NaCl + NH4HC03 =± NaHC03 + NH4C1.

This ammonia-soda process was worked on a technical scale by
Schloesing and Holland, from whose paper (1855) the following account
of the chemistry of the process is taken. Of the multitude of types of
apparatus described in Solvay's later patents, practically only the
c^bonating tower (p. 783) is still in use ; even this is not essential,
'j 9 ammonia -soda process was introduced by John Brunner and
Kobert Mond, in 1874, at Winnington, near Northwich, in Cheshire,
and in 1904 the works of Brunner, Mond and Co. converted 1,703,805
tons of salt into sodium carbonate. The Leblanc industry steadily
declined, and in 1908, out of a total world's production of 2 million
tons of soda, only 100,000 tons were made by the older process. Another
severe blow was given to the Leblanc process by the introduction of
electrolytic methods in 1895 (Chapter XVI).

The raw materials for this process are common salt (or brine),
limestone, coal, and ammonia. It consists of a cycle of six opera-
tions, which are carried on continuously day and night :

(1) A solution of salt is prepared, containing 31 per cent, of NaCl,
ammonia, and ammonium carbonate, freed from the calcium and
iron, and most of the magnesium, salts of the original brine.

(2) This ammoniacal brine is treated with carbon dioxide, which
first converts the ammonia into carbonate :

(i) 2NH3 + H20 + C02=± (NH4)2C03,
and then tends to convert this into bicarbonate :
(ii) (NH4)2C03 + H2O + C02 =± 2NH4-HC03.

In proportion as ammonium bicarbonate is formed it reacts with the
sodium chloride, giving by double decomposition sodium bicarbonate,
NaHC03, and ammonium chloride :

(iii) NH4-HC03 + NaCl ;=± NaHC03 + NH4C1.

The former salt is only slightly soluble in brine, and is nearly all
precipitated, whilst the latter remains in solution. Only two-thirds
of the common salt is converted into sodium bicarbonate, since the
reaction is reversible, and one-third of the salt and of the ammonium
bicarbonate remain. The mother liquor, which passes to the
ammonia-stills, therefore contains one-third of its ammonia
"volatile " and two-thirds " fixed " (p. 550).

(3) The bicarbonate is filtered, and washed so as to free it as far as
possible from ammonium salts.

(4) The bicarbonate is ignited to produce sodium carbonate and
nearly pure carbon dioxide (" roaster CO2").

(5.) The ammoniacal salt solutions from (2) and (3) are treated in

xxxvin THE METALS OF THE ALKALIES 783

stills with steam and lime to set free the ammonia, and form calcium
chloride.

(6) Limestone is burnt to produce carbon dioxide diluted with
nitrogen (" limekiln C02"), and the lime required for operation (5).

The products of the process are nearly pure sodium carbonate,
nitrogen containing a little carbon dioxide, and a solution of calcium
chloride. The two latter are waste -products.

A diagrammatic scheme of the ammonia-soda process is given below

SALT AMMONIA LIMESTONE COKE

1-65 tons 5 Ib. to make good loss 1| tons 1-3 cwt.

I !

Ammoniacal brine

— v-

Carbon dioxide -j- N2

)£ NITROGEN -H
Bicarbonate . Ammoniacal solution

\ /* from filters

COAL '

and
Lime

Carbon SODA- ASH Calcium chloride

dioxide 1 ton 1 ton

The operations are carried out as follows. The brine is saturated
with ammonia gas from the stills, and the precipitated impurities
(CaCOg, MgC03, Fe(OH)3) allowed to settle. The ammoniacal brine
is then pumped through pipes to the iron carbonating towers
(Fig. 384), 6ft. in diameter and 70-90ft. high, provided with perforated
inverted bubblers, as shown. Carbon dioxide (obtained by mixing
the roaster and limekiln gases) is pumped in below through the distri-
buter, 6, and bubbles through the ammoniacal brine in the tower,
forming a sludge of bicarbonate which runs off to the filters from e.
The liquor from the filters, containing all the ammonium salts,
passes to the ammonia-s'ills, where it is treated with steam and lime
in the usual way (p. 551), forming ammonia gas (with some carbon
dioxide), which is used in the preparation of ammoniacal brine, and a
solution of calcium chloride, which is a waste product (containing
the lime used, and the chlorine of the salt). The sodium bicarbonate
from the filters is then calcined in closed tubular calcining pans,
fitted with scrapers which push the solid along the pan. Carbon
dioxide is evolved : 2NaHC03 = Na2C03 + H2O + C02. This
gas is mixed with the scrubbed gas from the limekilns, where the
limestone is burnt mixed with coke (p. 841), and passed to the
carbonating towers. Sodium carbonate, or soda-ash, issues from

784

INORGANIC CHEMISTRY

CHAP.

1^>

FIG. 384.— Carbonating Tower of
Ammonia-Soda Process.

the calcining pan. This is nearly
pure ; it usually contains only a little
sodium chloride, derived from the mother
liquor left in the bicarbonate on the
filters.

From the soda-ash of the ammonia-
soda works, various products may be
made. Washing-soda, Na2C03,10H2O, is
obtained by dissolving in hot water
and crystallising. Crystal carbonate,
Na2C03,H2O, is -formed by evapora-
tion, and separates from the hot
solution. Concentrated soda crystals,
Na2CO3,NaHC03,2H2O, are produced
by crystallising a hot solution of equi-
molecular amounts of carbonate and
bicarbonate.

Caustic soda is made by boiling the
solution of the carbonate with lime, as
already described, or by the Lb'wig process.
In the latter, a mixture of soda-ash and
ferric oxide is heated to bright redness
in a revolving furnace, when sodium
ferrite, Na2O,Fe203, or NaFe02, is
formed :
NaaC08 + Fe2O3 = 2NaFe02 + C02.

The mass is cooled, broken up, and
thrown into hot water, when hydrolysis
of the ferrite, with formation of caustic
soda and insoluble ferric oxide, which is
used again, occurs :
2NaFe02 + H2O = Fe203 + 2NaOH-

The caustic soda solution is concen-
trated in vacuum evaporators, and
finally heated in soda-pots (p. 781) over
a free fire to produce fused caustic soda,
which is ladled into iron drums.

Sodium carbonate, Na2C03. — Anhy-
drous sodium carbonate, known as
soda-ash, is a white amorphous powder,
which aggregates on exposure to moist
air, owing to the formation of hydrates.
It melts at 852 °. When added to water,
a considerable amount of heat is
evolved, and the hydrated salt is formed

xxxvin THE METALS OF THE ALKALIES 785

usually as an agglomerated mass, which then dissolves slowly.
The solution is distinctly alkaline, owing to hydrolysis :
NaaC08 — 2Na + CO3" ; 003* + H2O — HC03' + OH'. In
Q-IN solution, 3-17 per cent, of the salt is hydrolysed, i.e.,
from every 100 molecules of Na2CO3 dissolved, 3-17 molecules
of NaOH are formed. The solution slowly loses carbon dioxide
on boiling.

On evaporating the solution, and cooling, large monoclinic
crystals of washing-soda, Na2C03,10H2O, are deposited. These
effloresce in the air, forming a white powder of the monohydrate
Na2C03,H2O, which is also formed from the decahydrate at 35-1°.
This form is deposited from hot solutions, and is known as crystal
carbonate ; it occurs native in the soda lakes of Egypt. Other
hydrates, e.g., Na2CO3,7H2O, are known.

Sodium bicarbonate, NaHC03. — This salt is produced in large
quantities by the ammonia-soda process, but is all converted into
carbonate, the bicarbonate of commerce being prepared from the
latter. A concentrated solution, or moist crystals, of sodium car-
bonate give, when saturated with carbon dioxide, a white crystalline
precipitate of bicarbonate. This may be washed with a little cold
water, in which it is sparingly soluble, and dried in the air :
CO/ + CO2 + H2O^2HC03'. The precipitation is due to the
fact that, in concentrated solutions, the solubility-product (p. 358),
[Na'J x [HCO3'J, of the salt is readily exceeded. The precipitated
bicarbonate is easily freed by washing from impurities contained in
the original carbonate (e.g., NaCl), since these are readily soluble,
and if it is gently ignited in a platinum crucible, pure sodium car-
bonate is produced, which may be used as a standard in volumetric
analysis : 2NaHCO3 ^± Na2C03 + H20 + C02. The solution of the
bicarbonate is slightly hydrolysed, and has an alkaline reaction,
though this is much feebler than that of the carbonate : HCO3' -f-
H2O ^ OH' -f H2C03. On heating the solution, bubbles of carbon
dioxide are evolved : H2CO3 ^ H2O + C02. By prolonged boiling,
practically all the bicarbonate is converted into carbonate, and if
crude bicarbonate from the ammonia-soda process is boiled with
water, the ammonium salts are expelled as well. On recarbonating,
almost pure sodium bicarbonate is precipitated, and the commercial
salt is made in this way.

Sodium sesquicarbonate, Na2C03,NaHC03,2H20. — If equimole-
cular amounts of sodium carbonate and sodium bicarbonate are
dissolved in warm water, and the solution cooled to 35°, monoclinic
crystals of sodium sesquicarbonate, Na2CO3,NaHCO3,2H2O, are
deposited. This salt occurs naturally as trona, or urao, in various
localities, and is produced by the spontaneous evaporation of soda
lakes. The artificial salt, known as concentrated soda crystals,
is used in wool-washing. It is neither efflorescent nor deliquescent.

786

INORGANIC CHEMISTRY

CHAP.

Large deposits of sesquicarbonate occur at Magadi, in British East
Africa, and are worked by the Magadi Soda Company.

Metallic sodium. — Although first prepared (Davy, 1807) from
caustic soda by electrolysis :

2NaOH

- 2Na

2Na
t

20H'

H2O

metallic sodium was for many years produced on the large scale
by a process due to Castner (1886). In this, caustic soda was
s ^n s z- heated to 1000° in

iron retorts with
crude carbide of iron,
prepared by heating
pitch with spongy
iron from the reduc-
tion of pyrites-cinder
with water gas :
6NaOH + 2C = 2Na +
3H2 + 2Na2C03. In
1891 Castner, on
account of the develop-
ments in the economi-
cal generation of
electricity, was able
to revert to Davy's
original process, and
all the sodium of
commerce (about 5000
tons per annum) is
now produced by this
method.

The electrolysis of sodium chloride, mixed with potassium or
calcium chloride, or sodium fluoride, is also carried out.

The caustic soda is fused in a cylindrical iron pot (Fig. 3C5), and
maintained at a temperature not higher than 330° by a ring of gas-
burners, g. A cylindrical iron cathode, h, passes up through the
base, and is sealed by solidified caustic soda, k, into a prolongation,
b, of the pot. The anode is a cylinder of nickel, /, and is in electrical
connection with a wire gauze cylinder, m, surrounding the cathode.
The metal rises from the cathode, and floats at d on the surface of the
caustic soda inside a small metal receptacle, c, provided with a lid, n.
It is removed by a wire gauze spoon, which allows the fused caustic
soda to flow away, but retains the sodium. The latter is sent out,

FIG. 385.— Manufacture of Sodium by Electrolysis.

xxxviii THE METALS OF THE ALKALIES 787

sealed up in tin cans, in the form of thick rods. This process is carried
out by the Castner-Kellner Co. at Newcastle, and at Clavaux (France) ;
by the Niagara Electrochemical Co. in America ; and by the Elektro-
chemische Fabrik Natrium at Rheinfelden, in Germany. The metal
is used in the preparation of cyanides, sodium peroxide, silicon, mag-
nesium, and or gano -metallic compounds in the dye industry.

Sodium is a silver-white soft metal, which may be obtained in
octahedral crystals on slow cooling of fused sodium. A colloidal
solution in ether has the same violet colour as the vapour. The
clean, freshly-cut surface of the metal rapidly tarnishes in the air,
a green phosphorescence being visible in the dark. The metal burns
when heated in oxygen or chlorine, but may be distilled unchanged
in the perfectly dry gases. It acts violently on water :
2Na -f 2H20 = 2NaOH + H2 (p. 181). Sodium is a good conductor
of electricity ; its conductivity is about 36 per cent, that of silver
(the best conductor).

Oxides of sodium. — Two oxides of sodium are known : sodium
monoxide, Na2O, a basic oxide, and sodium peroxide, Na2O2, or
Na-OO-Na.

Sodium monoxide is obtained either by burning sodium at 180° in
a limited supply of air or oxygen and distilling off the excess of metal
in a vacuum, or by heating sodium peroxide, nitrate, or nitrite with
sodium : 2NaNO3 + lONa = 6Na2O + N2. It is a white amorphous
mass, which decomposes at 400° into the peroxide and metal. It
reacts violently with water : Na2O -f H2O = 2NaOH, but does not
absorb carbon dioxide at the ordinary temperature.

Sodium peroxide, Na<.02, is produced when the metal burns in
excess of air or oxygen, and is manufactured by heating sodium in
aluminium trays in a current of purified air at 300° in iron pipes,
about 500 tons being produced annually. Sodium peroxide is a
yellow substance, becoming white on exposure to air from formation
of sodium hydroxide and bicarbonate. When very strongly heated,
it evolves oxygen. An aqueous solution may be prepared by adding
the powder in small quantities at a time to a well-stirred mixture
of ice and water, a crystalline hydrate, Na2O2,8H2O, being formed.
The liquid is strongly alkaline, owing to hydrolysis : Na202 -f- 2H2O
^± 2NaOH -f H2O2. On warming, oxygen is evolved. Carbon
dioxide decomposes sodium peroxide with evolution of oxygen,
hence the solid has been used for purifying air in confined spaces
(e.g., in submarines). The solution is an oxidising agent, e.g., it
converts chromic hydroxide into sodium chromate ; and the fused
salt shows powerful oxidising properties, converting chrome -
ironstone (FeO,Cr2O3) into ferric oxide and soluble sodium
chromate.

3 E 2

788 INORGANIC CHEMISTRY CHAP.

EXPT. 316. — A little sodium peroxide mixed with sawdust is placed
on filter-paper and moistened with water : the mass inflames. If
mixed with pieces of recently ignited charcoal and heated in a covered
porcelain crucible to 300—400°, a violent reaction occurs, and
metallic sodium condenses on the lid of the crucible : 3Na2O2 + 2C =
2Na2CO3 + 2Na. Glacial acetic acid inflames when the peroxide is
dropped into it.

If sodium peroxide is treated with absolute alcohol at 0°, a white
powder of sodyl hydroxide, or sodium hydrogen peroxide, Na-O-O-H,
is formed : Na2O2 + EtOH == NaOEt + NaO-OH. It explodes on
heating, evolving oxygen, and forming caustic soda. A stable com-
pound, 2NaO2H,H2O2, is formed on mixing 30 per cent, hydrogen
peroxide with sodium ethoxide (NaOEt) and absolute alcohol. By
the action of an ethereal solution of H2O2 on sodium, a white solid,
2NaHO2,H2O2, is obtained. Potassium forms 2KHO2,H2O2.

Sodium hydride, NaH. — Sodium hydride is prepared by passing
a slow stream of dry hydrogen over sodium in a nickel boat, heated
in a glass tube to 365°. Colourless matted crystals form on the
upper cooler portion of the tube just beyond the boat. These are
decomposed by water, with evolution of hydrogen : NaH + H2O =
NaOH + H2. Sodium hydride is not acted upon by concentrated
sulphuric acid. At 430° it dissociates rapidly: 2NaH^2Na +
H,,. It absorbs carbon dioxide, producing sodium formate :
NaH + C02 = Na-CO-OH.

Sodium cyanide, NaCN (or NaCy). — This salt is formed by adding
hydrocyanic acid to caustic soda : NaOH -|- HCN ^ NaCN -f-
H2O, and by heating sodium ferrocyanide, alone or with sodium :
Na4Fe(CN)6 = 4NaCN + Fe + 2C + N2 ;> Na4Fe(CN)6 + 2Na =
6NaCN -f- Fe. The pure cyanide is precipitated by passing hydro-
cyanic acid into an alcoholic solution of caustic soda. Sodium cyanide
is made on a large scale by Castner's process ; ammonia is passed
over sodium heated in iron retorts to 300-400°, and the fused
sodamide produced is poured over red-hot charcoal, when sodium
cyanamide, Na2:N-CN, is formed. This reacts with the heated
charcoal, forming sodium cyanide :

2Na + 2NH3 = 2NaNH2 + H2.

2NaNH2 + C - CN-N:Na2 + 2H2

CN-N:Na2+ C = 2NaCN.

Sodium cyanide is hydrolysed in aqueous solution ; the latter is
alkaline, and smells of hydrocyanic acid : NaCN -f H20 —
NaOH + HCN.

Sodium in analysis. — Sodium compounds give an intense yellow
flame when heated on platinum wire in the Bunsen flame. The light,
on examination by the spectroscope, shows two yellow lines, very

xxxvm THE METALS OF THE ALKALIES 789

close together, constituting what is known as the double D-line ;
their wave-lengths are 5896 and 5890 A.U. This spectrum is given
by practically every solid heated in the flame, since sodium is very
widely distributed in Nature. White, sparingly soluble, precipitates
of the sodium salts are formed when potassium pyroantimoniate
(p. 935) or potassium dihydroxytartrate are added to fairly
concentrated solutions of sodium compounds.

POTASSIUM, K = 38-79.

Potassium compounds. — Potassium occurs much less abundantly
than sodium, although it is widely distributed throughout the
three kingdoms of Nature. Primary rocks often contain potassium
silicate; thus granite contains 1-7 to 3-1 per cent, of potassium,
mainly in the form of orthoclase felspar (p. 891) : K2O,Ai2O3,6SiO2.
Potash mica, or muscovite, has the formula (KH)2Al3(Si04)3. During
the weathering of these rocks, i.e., their decomposition by atmospheric
carbon dioxide and water, assisted by the disintegrating action of
frost, the silicates are decomposed into clay and soluble potassium
salts, such as potassium carbonate. The latter are retained by a
process of adsorption in the soil, where they remain available for
absorption by the roots of plants (p. 696). The mechanism of the
selective retention of potassium salts by the soil appears to depend
on the exchange of potassium for sodium in zeolites, or hydrated
silicates :

Na2O,Al2O3,3SiO2,2H9O (natrolite) + 2KOH =
K20,Al2O3,3SiO"2,2H2O + 2NaOH.

In plants, potassium compounds occur as salts of organic
acids : e.g., sorrel and rhubarb contain acid potassium oxalate,
KHC204,H2C2O4,2H20, "salt of sorrel," or " salts of lemon,"
used to remove ink-stains from linen ; and grape- juice contains
acid potassium tartrate, KHC4O4O6, " cream of tartar," or " argol."
When plants are burnt, these organic salts form potassium carbonate,
K2C03, which, since it was formerly prepared by calcining cream of
tartar, received the name salt of tartar. Large amounts of potassium
carbonate are made in Canada, Transylvania, and Russia, by lixiviat-
ing wood ashes with water, evaporating the solution to dryness, and
calcining the residue in iron pots. The product is pot-ash ; when
purified it is known as pearlash.

According to Dyer (1894), the minimum amount of soluble potash
(K2O) in a fertile soil is 0-01 per cent. ; the mean available potash
content of British soils is 0-015 per cent. If successive crops are
grown on the soil, the potassium compounds are removed, and the
soil becomes infertile. Trees remove annually 1-25 Ib. of K2O per
acre, other plants more (p. 696). This exhaustion of the soil is occurring
in America, where the wheat-growing areas are moving further and

790 INORGANIC CHEMISTRY CHAP.

further west. In order to keep up the fertility of the soil, potassium
compounds must be supplied ; they are therefore essential as fertilisers.

The interesting suggestion has been made that the occurrence of
potassium compounds in plants, and the fact that the latter cannot
grow without potassium compounds, are connected with the feebly
radioactive properties of this element. The metal emits /3-rays, but
its activity is only one -thousandth that of uranium. Radioactive sub-
stances are said to promote plant growth even in the absence of
potassium salts.

Plants serve as food for animals, and the blood serum of all
animals contains 0-022 per cent, of potassium and 0-32 per cent,
of sodium. In the milk of carnivora, sodium and potassium occur
in approximately equivalent amounts ; in that of herbivora, and
in human milk, potassium predominates (6 : 1). The perspiration
of the sheep is rich in potassium salts of the organic sudoric acid.
If raw wool is washed with water, the brown liquid evaporated,
and the residue calcined, about 5 parts of potassium carbonate
remain per 100 of wool. This is a limited source of potassium salts.

Potassium salts occur in the sea, and are absorbed in marine
plants, from the ashes of which (kelp) they may be extracted. Sugar
beets absorb from the soil considerable amounts of potassium salts,
which accumulate in the molasses, known as vinasse, or schlempe.
They are evaporated on open hearths, and splashed by paddles in
the fire gases (Porion furnace) ; the syrup burns, leaving a residue
of potassium carbonate. The vinasse may also be distilled in iron
retorts, when methyl chloride and trimethylamine are formed.

Deposits of potassium salts. — Although potassium salts are widely
distributed, e.g., as felspar, comparatively few workable deposits
of salts occur. The principal deposits are found at Stassfurt, in
Saxony ; at Mulhouse, in Alsace ; at Cardona, in Spain ; and, in
lesser amounts, in Eastern Galicia, Searle's Lake (Nebraska), and
Elton Lake, in the Urals.

The Stassfurt potash deposits held, until quite recently, the
monopoly of the world's supply. They were discovered in boring
for rock-salt in 1839, and are of great thickness. The arrangement
of the deposits is as follows :

Top.

Alluvial and diluvial deposits.

" Bunter " sandstone— Triassic formation (600-800 ft. thick).

Gypsum, anhydrite, red clay, etc.

Newer common salt (a later formation, often lacking).

Anhydrite.

*' Salzthon " (three layers : bottom, of gypsum ; middle, of mag-
nesia and alumina ; top, of clay containing 40 per cent, of MgCO3, pro-
tecting the lower deposits).

xxxvin THE METALS OF THE ALKALIES 791

Carnallite region, chiefly KCl,MgCl2,6H2O (50-130 ft. thick).
Kieserite region (chiefly MgSO4,H2O) — " Abraurn " salts, i.e., above
common salt.

Polyhalite region — mixed salts.
Older common salt (2000 ft.).
Anhydrite.
Bituminous sandstone.

The deposits are probably derived from the evaporation of an
inland lake, as the order of the successive layers of salts is what
would be expected in such a case (Van't Hoff).

The chief source of potassium salts in the Stassfurt deposit is
the double salt carnallite, KCl.MgCl2,6H20, which contains, when
pure, 14-1 per cent, of potassium. The Alsatian and Galician
deposits contain sylvine, an isomorphous mixture of sodium and
potassium chlorides, richer in potassium than carnallite. The
kainite, K2S04,MgSO4,6H20, of Stassfurt is not worked to any
extent.

Potassium carbonate. — In the preparation of potassium salts
from carnallite, the latter may be fused, when nearly pure potassium
chloride separates, leaving fused hexahydrate of magnesium
chloride : KCl,MgCl2,6H20 =± KC1 + MgCl2,6H20. The potassium
chloride is recrystalfised. Usually, the carnallite is treated with
mother-liquor from the crystallisations. On heating the paste of
potassium chloride and the saturated solution of magnesium
chloride, formed by the action of water on the double salt, a clear
solution is obtained, from which on cooling 80 per cent, of the
potassium chloride is deposited. From the chloride the sulphate
and carbonate are prepared by a modification of the Leblanc pro-
cess. The charge for the black-ash furnace (p. 778) consists of
100 parts of K2S04, 80-90 parts of limestone, and 40-50 parts of coal.

Potassium chloride is also converted into carbonate by Precht's
process. A concentrated solution is. mixed with solid hydrated mag-
nesium carbonate, and carbon dioxide (limekiln gas) passed through.
A solid of the composition MgCO3,KHCO3,4H2O and a solution of
magnesium chloride are formed : 3(MgCO3,3H2O) + 2KC1 aq. + OO2 =
2(MgC03,KHC03,4H20) + MgCl2 aq.

This solid is heated to 140° with water under pressure. A solution
of potassium carbonate, a precipitate of magnesium carbonate, and
carbon dioxide gas are formed :
- 2(MgCO3,KHCO3,4H2O) = 2MgCO3 + K2CO3 + 9H2O + CO2.

Potassium carbonate, K2C03 (pearlash), is a white deliquescent
powder, dissolving readily in water to form a strongly alkaline
solution : K2C03 -j- H2O ^± KHC03 + KOH.

792 INORGANIC CHEMISTRY CHAP.

One hundred parts of the water dissolve :

0° 26° 40° 60° 80° 135° (b.-pt. sat. sol.)

K2CO3 105 113-5 117 127 140 205

It fuses at 879 °, but melts at a lower temperature when mixed
with sodium carbonate — fusion mixture, and loses carbon dioxide
when heated to redness in steam : K2CO3 + H2O = 2KOH + C02.

A crystalline hydrate, K2CO3,2H20, is stable in contact with
water from — 7° to 135°. The concentrated solution on standing
deposits monoclinic crystals of 2K2C03,3H2O, which at 100° fall
to a white powder of K2CO3,H20, and at 130° yield the anhydrous
salt.

Potassium carbonate solution readily absorbs carbon dioxide,
and a saturated solution on cooling deposits monoclinic crystals of
potassium hydrogen carbonate, or "potassium bicarbonate, "~KHC03
(or K20,2C02,H2O), which is easily prepared by passing carbon
dioxide over moistened potassium carbonate and drying on a
porous plate. The salt is much less soluble in water than the
normal carbonate. One hundred parts of water dissolve at 10°, 27 '7,
and at 60°, 60 parts of K2CO3.

The recrystallised bicarbonate may be used in preparing pure
potassium carbonate, since it decomposes at 190°: 2KHCO3^
K2CO3 -f- H2O + C02. The properties of the solution are similar
to those of sodium bicarbonate (p. 785).

Potassium hydroxide, or caustic potash, KOH.— Caustic potash is
prepared in a similar manner to caustic soda, which it resembles
closely in its properties (p. 781). It is made on Che large scale by
the electrolysis of a solution of potassium chloride, and is used
in the manufacture of soft-soap (potassium salts of oleic, palmitic,
and stearic acids). The pure hydroxide is prepared by the action
of barium hydroxide on potassium sulphate : K2S04 (powder) -f-
Ba(OH)2 (hot saturated solution) — BaS04 (pp.) + 2KOH, or
by the action of water on potassium amalgam. It forms a crys-
talline hydrate, KOH,2H20, m.-pt. 35-5°, although solutions con-
taining more than 85 per cent, deposit KOH on cooling. The
solutions attack glass, and should be decanted (not filtered), and
evaporated in silver, nickel, or iron dishes. Platinum is attacked
by fused alkalies.

Potassium chloride, KC1. — This salt occurs in cubic crystals as
sylvine. It melts at 790°, and is easily soluble in water, the solu-
bility increasing from 28 at 0° to 32-7 at 15° and 56-5 at 100°
almost linearly with temperature (cf. Fig. 68). The salt is made
from carnallite, as previously described, and is used as a fertiliser.

The bromide, KBr, and iodide, KI, are prepared as previously
described (p. 395) ; they form cubical crystals, which melt at
750° and 705° respectively, and are used in medicine and photo-
graphy. The fluorides, KF, KHF2, KH2F3, and KH3F4 are known.

xxxvni THE METALS OF THE ALKALIES 793

Potassium phosphate, K3PO4, is formed by heating a phosphate, or
phosphatic slag (p. 981), with coke and potassium sulphate, and is
used as a fertiliser.

A phosphide, K2P5, is obtained by heating the elements at 400° in
an exhausted tube. The compounds Na2P5, Rb2P5, and Cs2P5 are
similarly obtained.

The metaborate, KBO2, is prepared by fusing K2CO3 with B2O3 ;
on adding potash to boric acid till the solution is alkaline, a pyroborate,
K2B4O7,5H2O, is formed. By mixing H3BO3 and 2K2CO3 in hot solu-
tions, a triborate, 2KB3O5,5H2O, is formed, whilst the pentaborate,
KB5O8,4H2O, is made by saturating hot caustic potash solution with
boric acid.

By adding cold 3 per cent., and 30 per cent., H2O2, respectively, to
a saturated solution of the metaborate, two perborates, 2KBO3,H2O
and 2KBO3,H2O2, are obtained.

Potassium in analysis. — Potassium forms sparingly soluble salts
with perchloric, fluosilicic, chloroplatinic, tartaric, and picric acids,
all of which, together with sodium cobaltinitrite (p. 1001), may be
used as reagents for the potassium ion. The potassium hydrogen
tartrate is precipitated only in solutions containing no mineral
acid ; its precipitation (as well as that of the chloroplatinate) is
facilitated by adding alcohol, and scratching the tube with a glass
rod. The purple flame coloration and the spectrum are also
useful as tests.

Potassium cyanide, KCN. — Potassium cyanide is formed by heating
the ferrocyanide alone at a bright red heat : K4Fe(CN)6 =
4KCN + Fe -f 2C + N2, or with potassium carbonate : K4Fe(CN)6
+ K2CO3 = 5KCN + KCNO (cyanate) + C02 + Fe. If the ferro-
cyanide is fused with sodium, a mixture of sodium and potassium
cyanides is formed :

K4Fe(CN)6 + 2Na = 4KCN + 2NaCN + Fe.

Potassium cyanide is now prepared by Beilby's process ; a mixture
of fused potassium carbonate and carbon is treated with ammonia
gas : K2C03 + C + 2NH3 = 2KCN + 3H20. The fused cyanide
is decanted and moulded, and is quite pure. The cyanate, KCNO,
or CO:NK, is obtained by fusing the cyanide with lead oxide :
KCN -f- PbO = KCNO + Pb (the cyanide is a powerful reducing
agent), or by heating the ferrocyanide and potassium dichromate in
an iron dish, and extracting with 80 per cent, alcohol. The aqueous
solution slowly hydrolyses, with formation of ammonia :

KCNO + 2H20 - NH3 + KHC03.

The thiocyanate, KCNS, is formed by fusing a mixture of potass-
ium ferrocyanide and carbonate, with sulphur. It occurs in traces

794 INORGANIC CHEMISTRY CHAP.

in saliva. Potassium hydride, KH, is formed similarly to the
sodium compound.

Potassium. — The metal is prepared in a similar way to sodium
by the electrolysis of fused caustic potash.

EXPT. 317. — It may be obtained on a small scale by electrolysing
a fused mixture of equimolecular proportions of potassium chloride and
calcium chloride in a porcelain crucible, provided with two carbon
electrodes, and heating with a Bunsen burner placed on the anode side,
so that a solid crust forms over the cathode (Fig. 386). If six to eight
accumulators are used, a globule of potassium forms under the crust.
The whole is cooled, and opened up under petroleum.

Metallic potassium was formerly prepared
kj heating a mixture of the carbonate with
charcoal to whiteness in iron bottles, and
cooling the vapour rapidly in flat con-
densers: K2C03 + 20 = 2K + 3CO. Unless
the cooling was very rapid, combination of
potassium with carbon monoxide occurred,
with the formation of a yellow compound,
C606K6, which is a salt of hexahydroxy-
benzene, C6(OH)6. On exposure to moist
FIG. 386.— Preparation of air, this forms very explosive substances.

Potassium by Electro- 4^ , i i

lysis. Jrotassium can also be prepared by the

electrolysis of potassium cyanide, by heating

caustic potash or potassium sulphide with iron, magnesium, or
aluminium, or by heating calcium carbide with potassium fluoride.
It comes into the market in small spheres, preserved under petroleum.

Potassium is a very soft metal, with a silver-white colour. It is
not acted upon by perfectly dry oxygen, but is rapidly corroded in
moist air (c/. Na), becoming covered at first with a blue film. It
acts violently on water, the liberated hydrogen burning with a
purple flame (p. 775). When heated with practically every gas
containing oxygen, it abstracts the latter ; it also decomposes the
oxides of boron and silicon, and the chlorides of magnesium and
aluminium, on heating, with liberation of the elements. The metal
also occurs in traces in blue specimens of sylvine, which also contain
small quantities of helium and neon.

Oxides of potassium. — Potassium monoxide, K2O, is prepared in a
similar manner to Na2O (p. 787), and has similar properties. Potass-
ium tetroxide, K204, is obtained as a chrome-yellow solid by
burning the metal in oxygen or air (Gay-Lussac and Thenard), or
by the action of ozone on solid caustic potash : 2KOH -}- 03 =
K2O4 -f- H20. It oxidises carbon monoxide to dioxide below 100° ;
with water it forms H202, KOH, and oxygen. A dioxide, K2O2,
is said to be formed when K2O4 is exposed to moist air.

xxxvin THE METALS OF THE ALKALIES 795

Potassium and sodium sulphides. — Potassium and sodium burn
when heated in sulphur vapour, forming mixtures of sulphides. The
monosulphides, Na2S and K2S, are obtained by passing hydrogen
over the heated sulphates, and, in a less pure form, by heating the
sulphates with excess of carbon: K2SO4 -f 2C ="K2S -f 2CO2.
By fusing potassium carbonate with sulphur, a liver-coloured mass
is obtained, known as liver of sulphur (hepar sulphuris). It contains
polysulphides of potassium, together with potassium sulphate and
thiosulphate. A solution of liver of sulphur is used hi gardening
to combat mildew and insect pests.

If a solution of caustic potash or soda is saturated with
sulphuretted hydrogen, and evaporated, the hydrosulphides,
NaHS,2H20 or NaHS,3H2O, and 2KHS,H2O, crystallise out. The
anhydrous compounds are obtained by the action of sulphuretted
hydrogen on solutions of sodium or potassium in ethyl alcohol,
containing ethoxides. E.g., NaOC2H5 + H2S = NaHS -f C2H5-OH.

If to a solution of caustic potash or soda which has been saturated
with sulphuretted hydrogen an equal volume of alkali is added,
and the solution evaporated, the monosulphides, K2S,5H20 and
Na2S,9H20, separate in colourless crystals.

By boiling alcoholic solutions of the hydrosulphides with sulphur,
potassium pentasulphide, K2S5, and sodium tetrasulphide, Na2S4, are
obtained. K2S5 forms bright orange-red crystals, giving a deep
orange solution which becomes darker on heating. Na2S4 forms
dark yellow crystals, giving a deep orange solution which also
becomes darker on heating. Sodium disulphide, Na2S2, obtained by
adding sodium to an alcoholic solution of Na2S4, forms bright
yellow microscopic crystals, giving a deep yellow solution which
does-not darken on heating.

An examination of the freezing points of mixtures of the mono-
sulphides and sulphur (cf. p. 768) showed that the following sul-
phides exist :

K2S K2S2 K2S3 K2S4 K2S5 K2S6

Na2S Na2S2 Na2S3 Na2S4 Na2S5

K-S:S K-S:S

The constitution of these is probably : I , I , etc.

K-S K-S:S

LITHIUM, Li = 6-89.

Lithium.-— Lithium is a rare but widely distributed element.
It occurs in appreciable amounts only in a few rare minerals.
Traces of lithium are found in milk, blood, plants, especially
tobacco, and the soil. The lithium minerals are triphylite,
(Li,Na)3PO4 + (Fe,Mn)3(P04)2 (1 -6-3-7 per cent. Li); petalite,
LiAl(Si205)2 (2-7-3-7 per cent. Li); lepidolite, or lithium mica,

796 INORGANIC CHEMISTRY CHAP.

(Li,K,Na)2Al2(Si03)3(F,OH)2 ; and spodumene, LiAl(Si03)2 (3-8-5-6
per cent. Li). Lithium also occurs in the waters of certain mineral
springs, e.g., in Baden, and at Redruth, in Cornwall : in radioactive
minerals (e.g., carnotite) ; and in the sea. Traces of lithium are
found in most varieties of glass.

Lithium was discovered by Arfvedson (1817) in petalite and spodu-
mene : the metal was isolated by Bunsen and Matthiessen in 1855,
by the electrolysis of the fused chloride. Lithium may also be
obtained by the electrolysis of a solution of lithium chloride in
pyridine (C5H5N), and is a silver- white metal, harder than sodium,
tarnishing in the air, although less readily than the other alkali-
metals, and decomposing water, with evolution of hydrogen ; it
does not fuse on water like sodium and potassium, since its melting-
point is higher (180°).

Lithium salts are extracted from the minerals, such as spodumene,
in various ways. In one process the finely-powdered mineral is
digested with concentrated sulphuric or hydrochloric acid, which is
evaporated to render silica insoluble. The residue is taken up with
water, and the solution filtered. To the filtrate the requisite
amount of sodium carbonate is added to precipitate iron, alumina,
magnesia, etc., and the filtrate is concentrated by evaporation.
Excess of sodium carbonate is then added,, when lithium carbonate,
Li2CO3, is precipitated, as it differs from other alkali carbonates
in being sparingly soluble in water. Another process is to fuse the
mineral with barium chloride, extract with water, precipitate the
filtrate with baryta- water, and evaporate. The residue contains
sodium, potassium, and lithium chlorides, and is digested with a
mixture of absolute alcohol and ether, in which lithium chloride
alone is soluble. This salt (m.-pt. 606°) is one of the most deli-
quescent substances known.

Lithium burns, when heated in air above its melting point, with
a white flame, forming the monoxide (lithia), Li2O, a white substance
which dissolves slowly in water, with only moderate rise of tem-
perature, producing the hydroxide, LiOH. The latter is made by
decomposing an aqueous solution of lithium sulphate, Li2S04 (which,
unlike the sulphates of the other alkali-metals, is soluble in alcohol),
with baryta-water. It crystallises from the solution as LiOH,H2O,
and is a strong base. On heating the crystals in hydrogen at 140°,
a white porous mass of the hydroxide, LiOH, remains, and at 780°
the oxide, Li20, is formed. A peroxide, Li2O2, is formed by drying
over P2O5 the precipitate, Li2O2,H2O2,3H2O, obtained by adding
hydrogen peroxide and alcohol to a solution of the hydroxide.

Lithium carbonate, Li2CO3, and phosphate, Li3P04, are sparingly
soluble, and are precipitated from lithium chloride solution by the
corresponding sodium salts. The carbonate dissolves in a solution
of carbon dioxide, forming a solution of lithium bicarbonate, LiHCO3,

xxxvin THE METALS OF THE ALKALIES 797

which is more soluble than the normal carbonate (cf. CaHC03) .
The solution of the bicarbonate is called lithia water. On heating
the normal carbonate it decomposes completely into the oxide and
carbon dioxide. In these reactions, lithium shows a much closer
resemblance to the metals of the alkaline-earths, e.g., calcium, than
to those of the alkalies.

Lithium salts, especially those of organic acids (citrate, salicylate)
are used as a remedy for gout, since lithium urate is fairly soluble in
water (1 part in 368 parts of H20 at 20°). The nitrate, LiN03, is
very deliquescent, and soluble in alcohol.

Lithium salts give a splendid crimson flame when moistened
with hydrochloric acid and heated on a platinum wire in the Bunsen
flame. The light emitted is resolved by the spectroscope into a
very weak yellow line (6104 A.), and a brilliant crimson line (6708 A.).
Lithium is separated from potassium by the solubility of its chloro-
platinate, Li2PtCl6, and from sodium by the solubility of its chloride
in a mixture of absolute alcohol and ether, and in pyridine, in all of
which sodium chloride is insoluble. The sulphate, Li2SO4,H2O, is
readily soluble in water.

Lithium hydride, LiH, and lithium nitride, Li3N, are formed by
direct combination of the elements. The carbide, Li2C2, is formed
in the electric furnace, and with water evolves pure acetylene :
Li2C2 + 2H20 = 2LiOH + C2H2.

Rubidium and Caesium, Rb = 84-77, and Cs = 131-76.— Rubidium
and caesium occur in very small quantities in certain mineral waters
(e.g., Diirkheim, Ungemach, Bourbonne-les-Bains — 1 litre of the latter
contains 18-7 mgm. of RbCl and 32*5 mgm. of CsCl). Rubidium salts
are absorbed from the soil by plants, but caesium salts are not, and
act as vegetable poisons. These two elements were the first to be
discovered by the spectroscope (Bunsen, 1860). They give reddish-
violet and blue flame colours, respectively (Latin rubidus = darkest
red ; and ccesius = the blue colour of the sky). They also occur in
lepidolite, and some rare minerals. Carnallite (p. 791) contains about
0-035 per cent, of RbCl, which collects in the mother liquor from the
preparation of potassium chloride.

•These two elements may be separated from the other alkali -metals,
and from each other, by utilising the different solubilities of the chloro-
platinates and of the alums : —

Amounts in gm. of salts dissolved by 100 c.c. of water at 20° : —

K. Rb Cs.

Alums 13-5 2-27 0-619

R2PtCl6 1-12 0-141 0-070

Caesium carbonate is soluble in alcohol ; rubidium carbonate is
practically insoluble.

798 INORGANIC CHEMISTRY CHAP.

Rubidium salts are widely distributed, although in small amounts,
but caesium compounds are excessively rare. Although rubidium salts
are absorbed by plants, they cannot replace potassium, and the plants
die unless the latter is provided. Rubidium is feebly radioactive ;
its compounds emit /3-rays. The higher halogen compounds of
rubidium and caesium have been mentioned (p. 771).

AMMONIUM (NH4).

Ammonium compounds. — Ammonia, NH3, readily combines with
acids to form salt -like compounds. If a jar of hydrogen chloride is
inverted over one of ammonia gas, dense white fumes are produced
which settle on the sides of the jars as solid flakes of salammoniac,
NH4C1.

Lavoisier regarded these compounds as containing ammonia and
the acids ; on this view, which was extended by Dumas in 1828,
salammoniac would be ammonia hydrochloride, NH3,HC1. Ampere
(1818), however, supported the theory (first put forward by Davy in
1810) that these salts contain a radical ammonium, NH4, which
behaves as an alkali-metal. Salammoniac is therefore ammonium
chloride, NH4C1, analogous to potassium chloride, KC1. This view
of the constitution of the salts was favoured by Berzelius (1820).

The ammonium theory really had its origin in the discovery of
what is called ammonium amalgam, obtained independently by
Seebeck, in Jena, and by Berzelius and Pontin, in Stockholm (1808).
If a solution of ammonium chloride is electrolysed with a mercury
cathode (Fig. 153), the latter swells up in a curious manner, forming
a soft, pasty mass, which rapidly decomposes, evolving hydrogen
and ammonia in the ratio of 1 vol. to 2. This indicates that the
decomposition : N2H8 = H2 -j- 2NH3 has occurred. Davy (1810)
confirmed this observation, and showed that the " ammonium
amalgam " could also be obtained by the action of potassium
amalgam on a solution of ammonium chloride : 2K 4- 2NH4C1
= 2KC1 + N2H8.

EXPT. 318. — Add a little sodium amalgam to a cold solution of
ammonium chloride. Notice the way in which the amalgam swells
up. Place a little of the ammonium amalgam in water : bubbles of
hydrogen are evolved, and the liquid smells of ammonia.

Seely (1870) found by compressing ammonium amalgam in a
tube under a piston that it obeyed Boyle's law, and concluded
that it was simply a froth of hydrogen and ammonia gases in
mercury. Pfeil and Lippman found that a similar amalgam was
formed from salts of methyl amine, e.g., N(CH3)4Cl — methylamnK
nium chloride, whereas aniline salts, containing liquid aniline,
C6H6NH2, did not react. The methylamine salts can give the

xxxvin THE METALS OF THE ALKALIES 799

gaseous free base, N(CH3)3. These experiments tell against the
existence of free ammonium.

Other experiments speak in favour of the existence of ammonium
in the amalgam. Although the latter does not reduce solutions of
ferric chloride or copper sulphate at the ordinary temperature, it
reduces solutions of copper, cadmium, zinc, and even barium, salts
at 0°. The voltage required to deposit sodium on a mercury cathode
is similar to that required in the formation of ammonium
amalgam. An amalgam can be prepared by electrolysing a solution
of tetramethylammonium chloride, a substituted ammonium salt,
N(CH3)4C1, in absolute alcohol at 0°, with a mercury cathode. This
may contain N(CH3)4 or N2(CH3)8 ; it reduces copper and zinc salts
in alcoholic solution. The deep blue solutions obtained by dissolving
sodium or potassium in liquid ammonia may be metal-ammoniums,
NH3Na and NH3K, or N2HflNa2 and N2H6K2, or merely colloidal
solutions of the metals ; the latter can be filtered out under pressure.

Although there is no doubt as to the existence of the ammonium
ion, NH4fl, in solutions of ammonium salts, there is not yet con-
clusive evidence that the electrically neutral ammonium radical,
NH4 or N2H8, can exist in the free state.

Ammonium chloride, NH4C1. — This compound, known as salam-
moniac, is prepared by neutralising ammonia solution .with hydro-
chloric acid and 'evaporating. It is also made by boiling a solution
of ammonium sulphate, which is the commonest ammonium salt
(p. 552), with an equivalent amount of common salt : (NH4)2SO4
+ 2NaCl ^± Na2S04 + 2NH4C1. The sodium sulphate separates, and
is fished out : on cooling, ammonium chloride crystallises. It is
purified by recry stall isation, on by sublimation. T?he latter opera-
tion is carried out by heating the salt in a cast iron basin provided
with an iron dome, having a small hole at the top. The cake of
ammonium chloride which sublimes into the dome is broken up, and
forms tough, fibrous, irregular lumps, often stained in yellow patches
with ferric chloride. A mixture of ammonium sulphate and common
salt may also be heated in the same apparatus. An imitation of the
sublimed product is made by strongly compressing the powdered
salt : the wrell-known voltoids, used in batteries, are small tablets
prepared by compression. Ammonium chloride is prepared in
ammonia-soda works by crystallising the liquors from the bicar-
bonate filters, which contain NH4C1, NaCl, and CaCl2, and drying
the salt with hot air.

Ammonium chloride crystallises in feathery growths, consisting
of aggregates of small octahedra or other forms of the regular system
(p. 436), so that the crystals appear to belong to the hexago-
nal or tetragonal system. From a solution containing urea it
crystallises in cubes isomorphous with NaCl and KC1.

The salt is readily soluble in water, and a considerable lowering of

800 INORGANIC CHEMISTRY CHAP.

temperature results. It is very sparingly soluble in absolute alcohol.
The aqueous solution is only slightly hydrolysed, and is neutral, but
on boiling, ammonia escapes, leaving a distinctly acid liquid :
NH4C1 + H20 ±=; NH4-OH + HC1 ^ NH3 + H20 + HC1. Ammo-
nium chloride vapour is almost completely dissociated (p. 151) :
NH4C1 ^ NH3 -f HC1, unless the salt has been carefully dried over
P2O5, when it gives the normal vapour density corresponding with
NH4C1. The ready dissociation of the salt on heating explains
its action as a flux in soldering (p. 864) : the oxides are converted
into volatile chlorides by the hydrochloric acid, and a clean metal
surface is left.

Ammonium fluoride, NH4F, bromide, NH4Br, and iodide, NH4I, are
obtained by neutralising the corresponding acids with ammonia. The
salt NH4F,HF, is also known.

Ammonium sulphides. — If ammonia gas and sulphuretted hydro-
gen are mixed in proper proportions and the mixture is cooled,
ammonium sulphide, (NH4)2S, crystallises out. If equal volumes of
the gases are mixed at the ordinary temperature, solid ammonium
hydrosulphide, NH4-HS, is deposited. Both these compounds are
colourless : they dissociate on heating into NH3 and H2S.

If sulphuretted hydrogen is passed through concentrated ammonia
solution diluted with four times its volume of water, a solution of the
hydrosulphide is formed. The normal sulphide does not appear to
exist in solution. On cooling concentrated ammonia solution which
has been treated with sulphuretted hydrogen, crystals of compounds
of NH4-HS and (NH4)2S separate.

The freshly-prepared solution of the hydrosulphide is colourless,
but oxidises rapidly on exposure to air and becomes yellow, owing to
separation of sulphur, which dissolves in the excess of hydrosulphide
to form yellow polysulphides, (NH4)2Sa;. The same yellow ammonium
sulphide is obtained by digesting flowers of sulphur with the solution
of the hydrosulphide : the main product appears to be (NH4)2S4.
By distilling a dry mixture of salammoniac, quicklime, and sulphur,
a blood-red liquid is obtained containing polysulphides of ammonium
(NH4)2S*. The composition of these, which may be similar to the
potassium and sodium compounds (p. 795), has not been satisfac-
torily ascertained. On prolonged exposure to air, the solutions
deposit sulphur, and form a colourless solution of ammonium
thiosulphate, (NH4)2S203.

Ammonium sulphates, (NH4)HS04 and (NH4)2S04. — The manu-
facture of ammonium sulphate, (NH4)2S04, from ammonia has been
described (p. 552). Instead of using sulphuric acid as absorbent, a
German patent specifies the absorption of ammonia in a suspension
of calcium sulphate (gypsum), carbon dioxide being passed through
the liquid. Calcium carbonate is precipitated, and a solution of

xxxviii THE METALS OF THE ALKALIES 801

ammonium sulphate is formed : CaSO4 -f 2NH3 -f- C02 -}- H20 =
(NH4)2SO4 -f- CaC03. Ammonium sulphate when pure forms large
transparent crystals isomorphous with potassium sulphate (p. 512),
and very soluble in water. On heating they decompose, partly
with reduction to sulphur dioxide, nitrogen, and sulphur, and partly
with evolution of ammonia, and formation of the acid sulphate :
(NH4)2S04 = NH4-HS04 + NH3. This may also be obtained in
deliquescent crystals, by adding sulphuric acid to a solution of the
normal sulphate, and crystallising. The sulphite, (NH4)2SO3, is
obtained in crystals by passing sulphur dioxide through ammonia.
Ammonium nitrate, NH4N03. — This salt was first prepared
by Glauber, and was called nitrum flammans. It is obtained by
neutralising dilute nitric acid with ammonia or ammonium carbonate.
On the large scale it is made by passing ammonia gas into 60 per
cent, nitric acid ; by the double decomposition of calcium nitrate
and ammonium carbonate or sulphate ; by the double decomposi-
tion of ammonium sulphate and sodium nitrate : (NH4)2SO4 -f
2NaN03 ^± 2NH4N03 + Na2SO4 ; or by using sodium nitrate
instead of common salt in the ammonia-so/da process (p. 782) :
NaN03 + NH4-HC03^±NH4N03 + NaHC03. A direct method
of preparation has been described (p. 576).

The salt exists in five crystalline forms, with definite transition tem-

-17° 32'1° 84-2°

peratures : Tetragonal ^ (Rhombic)! ;=± (Rhombic)2 ^± Rhombo-

125-2" 169-6°

hedral ;zr Cubic ^ Liquid. The melting-point of the ordinary salt,
containing a little moisture, is 165°. The transition at 84-2° is accom-
panied by an expansion, which may break a glass vessel in which the
salt has solidified.

Ammonium nitrate is used in the preparation of nitrous oxide
(p. 582), and is also a constituent of explosives. A mixture of 80
parts of ammonium nitrate and 20 parts of trinitrotoluene (amatol)
was extensively used in the late war.

Ammonium nitrite is obtained as an explosive, deliquescent solid by
passing the red fumes from nitric acid and arsenious oxide (p. 587)
through lumps of solid ammonium carbonate in a cooled tube, dissolving
in alcohol, and precipitating with ether. It is formed by mixing the red
fumes with ammonia gas, as a white powder, although ammonium
nitrate is also produced.

Ammonium carbonates. — The preparation of commercial ammo-
nium carbonate, sal volatile, by the distillation of bones, horns, etc.,
was described by the later alchemists. The different materials were
supposed at first to yield different kinds of volatile alkali ; a particu-
larly valuable variety from the medicinal point of view was obtained
by distilling human skulls, especially of persons who had been

3 F

802 INORGANIC CHEMISTRY CHAP.

hanged, although the dry distillation of vipers furnished a product
which was also highly esteemed.

The salt is now obtained by a method described by Basil Valentine,
viz., by subliming a mixture of 2 parts of chalk and 1 part of salam-
moniac, or ammonium sulphate, in iron retorts with lead receivers.
The product is resublimed after the addition of a little water, and
conies into the market as a white, semi-transparent, fibrous mass,
covered on the outside with a white, opaque powder of the bicar-
bonate, NH4-HC03, and smelling strongly of ammonia. The
commercial carbonate is a mixture of the bicarbonate and ammonium
carbamate, NH4-C02-NH2. If the solid is treated with alcohol
the carbamate is dissolved, leaving the bicarbonate ; if it is exposed
to air, the carbamate slowly volatilises ; NH4-C02-NH2 ^r 2NH3 -f
CO2, leaving the bicarbonate as a white powder. The bicarbonate
can be crystallised ; at 60° it decomposes : NH4-HC03 ^±NH3 -f-
CO2 + H2O, although at the ordinary temperature it does not smell
of ammonia. Commercial ammonium carbonate can be used as
a baking powder since it volatilises completely on heating. If com-
mercial ammonium carbonate is treated at 30° with concentrated
ammonia solution, a sesquicarbonate, 2NH4HC03,(NH4)2CO3,H2O,
is obtained in crystals. The normal carbonate, (NH4)2C03, is obtained
by treating sal volatile with a small quantity of water, or by digesting
it for two hours with concentrated aqueous ammonia at 12°, and
drying the crystalline powder remaining, (NH4)2C03,H2O, between
filter-paper. It is formed when the carbamate is dissolved in
water :

C0< + H0 — C0

so that when the commercial carbonate is dissolved in ammonia
solution the normal carbonate is formed. The carbamate, is deposited
when 2 vols. of ammonia- gas and 1 vol. of C09 are mixed:
2NH3 + CO2 =± NH4-C02-NH2.

EXERCISES ON CHAPTER XXXVIII

1. Give a brief account of the views previously held on the nature
of the alkalies. How was Black able to demonstrate that the early
views were incorrect ? How were the alkali -metals isolated ?

2. How do potassium and lithium occur in Nature ? From what
sources, and by what methods, are potassium salts prepared on the
large scale ?

3. How are sodium carbonate and caustic soda manufactured from
common salt ? What takes place when a solution of sodium carbonate
is boiled with slaked lime ?

4. How may sodium and potassium compounds be differentiated
from each other in analysis ? If you were given a mixture of sodium

xxxvm THE METALS OF THE ALKALIES 803

chloride and potassium carbonate, how would you determine the
amounts of each salt present ?

5. Give an account of the Ammonia Soda Process. How may it
be modified so as to produce ammonium nitrate from Chile saltpetre ?

6. How are lithium salts prepared ? How do they differ from salts
of sodium and potassium ?

7. Where do rubidium and caesium salts occur ? How may these
two elements be separated ?

8. Why are ammonium salts grouped with those of the alkali-metals ?
What evidence is there of the existence of free ammonium ?

9. Describe the preparation and properties of : potassium iodide ;
ammonium carbonate ; ammonium nitrate ; sodium sulphides ;
potassium percarbonate.

10. Give a general account of the group of alkali-metals, paying
particular attention to the gradation of the properties of the elements
and their compounds with increase of atomic weight.

3F 2

CHAPTER XXXIX

COPPER, SILVER, AND GOLD

General properties of the group. — The metals of this group, which
occur in Nature in the free state, or else are very easily formed by
the reduction of their compounds, were the earliest known elements.
Although they occur in the same group as the alkali -metals, they
differ considerably from the latter ; the sole similarity is the exist-
ence of a series of compounds MX, hi which the metals are univalent.
This is the only type of combination known with silver, but copper
forms a series of compounds in which it is bivalent, CuX2, and gold
a series in which it is tervalent, AuX3, and both these are more
stable, and better known, than the univalent series. Unlike the
alkali-metals, copper, silver, and gold readily form complex com-
pounds, in which the metal may be present either in the positive
radical, e.g., [Cu(NH3)4]S04, or in the negative ra'dical, e.g.,
K[Ag(ON),]. .

Gold, having the highest atomic weight, differs in many respects
from the other members of the group ; this anomalous behaviour
occurs frequently in the periodic system. Gold in many ways
resembles platinum. Copper also shows a much closer relationship
with mercury, which forms a unvailent and bivalent series of com-
pounds (p. 870), than with silver or gold, although the cuprous salts
resemble those of silver. Cuprous and silver chlorides are both
white, insoluble substances, dissolving readily in ammonia. Although
silver chloride is quite stable, cuprous chloride is readily oxidised to
the cupric compound. The sulphides of copper and silver are
isomorphous ; the mineral copper glance, consisting chiefly of
cuprous sulphide, Cu2S, contains silver sulphide, Ag2S, in isomor-
phous admixture in varying amounts.

The heats of formation of some compounds, in kgm. cals., are given
below : the numbers for potassium are given for comparison :

B = K. Cu. Ag. Au.

R +C1 ... 104-3 32-85 29-4 5-8

R +Br ... 95-1 25-0 22-7 -0-1

R +1 ... 80-1 16-25 13-8 -5-5

R2 + O ... 164-6 40-8 6-9 ?

804

CH. xxxix COPPER, SILVER, AND GOLD 805

These values correspond closely with the affinities of the various
elements, since it has been shown by Nernst that in the case of solid
compounds the heat of formation is an approximate measure of the
affinity of the elements, although this does not usually hold for the
formation of gaseous or dissolved substances.

COPPER. Cu = 63-07.

Copper. — Copper occurs in the native, or metallic state, and was
therefore used in very early times, especially in the form of its
alloy bronze, which contains copper and tin. Working in bronze
was practised at least as early as 2000 B.C. ; the Bronze Age
succeeded the Stone Age, and preceded that of Iron.

Copper was originally obtained by the Greeks and Romans from
the island of Cyprus ; the Latin name aes cyprium or Cyprian brass,
afterwards became simply cyprium, and finally cuprum. These
names were, however, with the Greek chalkos, also used for brass
and bronze. The alchemists called the metal Venus (from its occur-
rence in Cyprus), and designated it by the symbol of the mirror, ? .
The precipitation of copper from the drainage-water of copper mines,
by iron, was considered to be a case of transmutation until Van
Helmont pointed out that the liquid originally contained a salt of
copper, derived from copper pyrites in the mine. Boyle (1675)
explained the reaction as one of simple displacement.

Native copper occurs in masses, and in veins traversing sandstone
in Sweden, the Ural mountains, and in large quantities in the vicinity
of Lake Superior. Cuprous oxide, Cu20, occurs as cuprite (or red
copper ore) ; cupric oxide, CuO, occurs in smaller amounts as tenorite,
or melaconite. Compounds of the carbonate and hydroxide occurring
native, especially in the Ural districts, are malachite, CuC03,Cu(OH)2,
and azurite (or chessylite), 2CuC03,Cu(OH)2, which are bright green
and deep blue in colour, respectively, and are used in works of art.
In combination with sulphur alone, copper is widely distributed,
although in relatively small amounts, in the forms of chalcocite, or
copper glance, Cu2S ; and covelline, CuS, both probably formed by
reduction of the sulphate by organic matter. The commonest
ores are copper pyrites, or chalcopyrite, CuFeS2, and erubescite
(or variegated copper ore), Cu3FeS3, i.e., sulphides of copper and iron.
Considerable quantities of copper are extracted by the " wet
process " from the residues left after burning iron pyrites containing
copper (cupreous pyrites), in the manufacture of sulphuric acid (p. 778).

Copper occurs in the red colouring matter of the feathers of the
toucan, and in the hcemocyanin of the blood of the cuttlefish, which
acts like haemoglobin (p. 697) as an oxygen carrier, but is blue in arterial
and colourless in venous blood. Minute quantities occur in plants,
especially in green peas.

806 INORGANIC CHEMISTRY CHAP.

Ordinary bread contains 4 mgm. of Cu per kgm., potatoes 2 mgm.
As much as 100 mgm. of copper may be taken with food per day without
danger, and higher organisms appear to have become to a certain
extent immune to copper, although traces of lead and mercury are
poisonous. Lower organisms, on the other hand, are very sensitive
to copper salts. Traces of the latter are added to drinking water in
America, to destroy bacilli and algae, and a solution of copper sulphate
mixed with chalk is used, under the name of Bordeaux mixture, for
spraying potatoes, etc., to prevent the growth of blight. Seed-corn
may also be steeped in a 0-5 per cent, solution of copper sulphate to
prevent the development of smut.

•

The annual production of copper amounts to about 1,000,000
tons ; about 650,000 tons were smelted in the United States in 1915.

Copper smelting. — Native copper is simply melted with a flux and
then refined. Oxides (e.g., cuprite) and carbonates (e.g., malachite)
are reduced by heating with carbon. Sulphide ores, from which a
large amount of copper is obtained, are smelted by a somewhat
complicated process, either in reverberatory furnaces (Welsh process),
or in the blast furnace (Mansfeld process) .

The Welsh process. — The simultaneous separation of the iron and
sulphur from the ore is a matter of difficulty, since sulphur has a
much greater affinity for copper than for iron. On roasting the ore,
the iron is mainly oxidised to ferrosoferric oxide, Fe304, whilst the
sulphur remains combined with the copper as cuprous sulphide,
Cu2S. This preliminary roasting is carried out in large flat furnaces,
the ore being raked on the hearth by mechanical means so as to
expose a large surface to the oxidising action of the air. The roasted
ore is then fused at a high temperature in a reverberatory furnace
with material containing silica. This combines with the oxide of
iron to form a readily fusible silicate of iron, whilst the cuprous
sulphide forms a lower layer, still containing some iron, called
coarse metal. Fig. 387 shows the section of a reverberatory
furnace used for copper smelting. The flames from the gas producer,
A, strike against the arched roof of the furnace and are deflected
on to the charge on the hearth, E. The secondary air for the com-
bustion of the gas enters through the holes, b, b ; that for the oxida-
tion of the charge is admitted through the ports, K, K. The
process is repeated, and nearly pure cuprous sulphide, called white, or
fine, metal, is obtained. Blocks of fine metal are then roasted on the
hearth of a reverberatory furnace, with a free supply of air. The
sulphur is partly burnt off, with formation of cuprous oxide, Cu20.
Reaction then takes place between the cuprous oxide and cuprous
sulphide, with formation of metallic copper : Cu2S + 2Cu20 =
6Cu -}- SO2. This roasting is carried out slowly ; the blocks retain
their shape, but become covered with blisters, due to escape of gas.

COPPER, SILVER, AND GOLD

807

XXXIX

This blister-copper still contains 2 to 3 per cent, of impurities, mainly
sulphur and iron. It is purified by melting a large quantity on a
furnace hearth, skimming off the slag, and then removing the oxygen,
dissolved in the metal in the form of cuprous oxide which would
render the copper brittle, by covering the surface of the metal with
powdered anthracite, and stirring with a pole of green birch-wood.
Torrents of reducing gases bubble up through the metal, and the
oxygen is removed. The metal is then tested by casting a small
ingot, which is half cut through with a chisel and broken off. If
the metal is sufficiently tough, the whole is cast in iron moulds.
If the reduction has been carried too far, the metal becomes brittle,
and is said to
be over-poled.
It is then ex-
posed to the
air for a short
time to allow
it to recover
its tough pitch
before casting.
It will be
seen from this
d e s c r i ption,
which does not
include all the
actual oper-
ations, that
the Welsh
method of
smelting cop-
per is a com-
plicated pro-
cess. The
effect of over-

FiG. 387. — Reverberatory Furnace.

poling may be due to the reduction of oxides of other metals,
which alloy with the copper and render it brittle.

The Mansfeld process. — In this method the ore is smelted in
blast furnaces, constructed of iron with a water cooling- jacket and
lined in the lower portion with firebricks (Fig. 388). The roasted
ore is mixed with coke or anthracite and a material containing silica,
and charged into the top of the furnace. Air is forced in through
pipes, 7, /, and reactions occur leading to the formation of a slag
and a matte corresponding with the coarse metal of the Welsh process.
The slag and matte flow into the fore-hearth, W, the slag running
away continuously from the opening, M, and the matte being
tapped from the hole, O, as required. The matte is poured into a

808

INORGANIC CHEMISTRY

CHAP.

rectangular Bessemer converter (cf. p. 979), and a current of air is
forced through it. The same reactions occur as in the former
process and copper is produced. Sulphur is burnt off as sulphur
dioxide, iron passes into the slag as silicate, and arsenic, etc., sublime.
Recent practice aims at smelting sulphide ores by the heat
of combustion of the sulphur in them, with the addition of about
5 per cent, of fuel, in rectangular blast furnaces provided with a
number of blowing pipes, or tuyeres. If a little boron is added
to the fused copper it combines with the oxygen, nitrogen, and
sulphur dioxide, and the cast metal is free from blow-holes, which
would result from the escape of these gases on cooling. The boron

is added in the form
of an alloy with copper.
Copper is also ex-
tracted by the wet
process. The ore is
leached with a solution
of ferric sulphate, and
a solution of copper
sulphate is obtained.
This is reduced by
metallic iron. The
burnt pyrites from the
manufacture of sul-
phuric acid, if they
contain copper, are
worked up by roasting
with 10-15 per cent,
of salt in large shelf
furnaces. The copper
chloride, CuCl2, formed

FIG. 388.— Blast Furnace for Manufacture of Copper. 1S extracted with

water, and any silver

and gold present are precipitated as iodides. The copper is then
reduced by scrap iron. In the Rio Tinto process, heaps of 100,000
tons of pyrites are exposed to air and rain. Slow oxidation occurs,
and the copper sulphate formed is washed out with water. The
remaining pyrites are exported for burning.

Copper refining. — Copper is largely used in the manufacture of
wires and cables for carrying electric currents, and since its con-
ductivity is appreciably lowered by traces of impurities, it is neces-
sary to use a highly purified metal. The traces of silver and gold
found in the crude metal are also of value. In refining copper, the
electrolytic process is exclusively used ; the large slabs of crude
metal are immersed in a bath of copper sulphate solution acidified
with sulphuric acid, and made the positive electrodes, or anodes

xxxix COPPER, SILVER, AND GOLD 809

in the bath. The cathodes consist of thin sheets of pure copper
covered with a layer of graphite (Fig. 389). The copper dissolves
from the anode as cupric ions, Cu", and these travel to the cathode,
where they give up their charges and are deposited as pure copper.
Iron and zinc pass into solution as sulphates ; gold and silver (with
some impurities) fall to the bottom as an anode slime, which is
collected and cupelled (p. 819) for the purpose of obtaining the
precious metals.

A similar process has been used since its invention by Jacobi in 1839,
in electrotyping, i.e., depositing copper electrolytically. This is used
in reproducing statues and other works of art. The copper may be
deposited on plaster casts covered with graphite to render them con-
ducting, arid the shell stripped off. In the same way, if an impression
of printers' type is taken on gutta percha, and the latter covered
with powdered graphite, a thin deposit of copper may be formed over

CURRENT OUT
CURRENT IN

-—PURE COPPER
UWJICOPPER

=
-

FIG. 389. — Purification of Copper by Electrolysis.

the surface by electrolysis. This is stripped off, and backed by pouring
on molten type-metal. The plate may then be used for printing. Copper
may be deposited on iron by dipping the metal in a solution of copper
cyanide in potassium cyanide, when a thin adherent film of copper is
deposited (a spongy deposit is produced frvom copper sulphate) ; this
is then thickened by electrolysis in a solution of copper sulphate. Iron
rollers are in this way covered with copper for use in calico-printing.

Copper is used for the driving-bands of steel projectiles. The driving-
band consists of a copper band recessed into a groove in the base of the
shell, and projecting slightly above the surface of the latter so as to be
somewhat larger than the bore of the gun. On firing the shell, the
copper is squeezed into the spiral rifling of the gun-barrel, and the
gases are prevented from escaping, whilst the shell acquires a rotation
which serves to keep it in its trajectory without turning over.

Alloys of copper. — The alloys of copper with other metals are of

810 INORGANIC CHEMISTRY CHAP.

great technical importance. Brass (copper -f- zinc) and bronze
(copper -f- tin) have been known from the earliest periods. They
were made by heating copper with zinc and tin oxides, in presence
of carbon. The tin or zinc oxide is reduced, and the metal alloys
with the copper. These alloys are now made by fusing the copper,
and adding the requisite amount of zinc or tin.

Copper. Tin. Zinc. Iron. Phosphorus.

Common brass ... 2 1

Bronze (gun-metal) ... 9 1

Speculum metal ... 2 1

Bell metal 4-5 1

Phosphor-bronze ... 94-75-82-5 5-15 - . 0-25-2-5

Delta metal 55 41 4

Dutch metal 80 20

Muntz metal 60 40

Old Roman coin ... 96-06 2-71 0-85

Modern bronze coin ... 95 4 1

Phosphor-bronze is hard, elastic, and tough ; delta metal can be
forged and rolled as well as cast, and is used for bearings, valves,
and ships' propellers. Muntz metal is used as a sheathing for
wooden ships. The definite compounds Cu3Sn and Cu4Sn are
known.

Properties of copper. — A new surface of copper appears light
red in colour, but this is due to the admixture of unchanged reflected
light with that from which parts have been abstracted by reflection
from the metal surface. The true colour of copper, produced by
selective reflection, is a deep rose-colour, as is seen by looking at
the fold of a piece of copper foil, cleaned with nitric acid, bent to a
V -shape. The light is then reflected many times from the surface
of the metal before entering the eye. The complementary colour,
green, is seen in the light transmitted through thin leaves of the
metal. Fused copper also emits a green light at high temperatures.

Pure copper is very malleable and ductile, and can be rolled into
sheets, hammered into thin leaves, and drawn into wire. The
metal may also be " spun " on the lathe, in the production of
seamless vessels. Just below the melting-point copper becomes
brittle, and appears to undergo allotropic change. Small quantities
of impurities reduce the malleability of the metal.

Pure electrolytic copper has a density of 8 -945 ; after hammering
or rolling the density increases to 8-95. Pure copper powder is
produced by allowing zinc to dissolve in a slightly acidified solution
of copper sulphate, washing the precipitated copper with hot water
and alcohol, and removing the small quantity of occluded hydrogen
by heating in a vacuum.

xxxix COPPER, SILVER, AND GOLD 811

The melting point of pure copper is 1083° ; the metal boils at
2310°, and can be distilled in a vacuum. The spongy and the fused
metals occlude various gases ; when the metal solidifies these form
bubbles, or give rise to " spitting " (p. 823).

On striking an arc under water between iron wires coated with
copper, a colloidal solution is obtained, but this probably contains
the oxide. By dialysing a solution of copper sulphate containing
sodium hydroxide, and sodium lysalbate or protalbate, and then
reducing, by warming with hydrazine, a dark red solution of colloidal
copper is produced. If only partially reduced, a yellowish-red
colloidal solution of cuprous oxide is obtained.

In the air, copper rapidly tarnishes, becoming covered with a
very thin adherent brown film of oxide or sulphide, which causes
the bright rose colour of the metal to deepen to brown. On pro-
longed exposure to moist air, a green film of basic carbonate (verdi-
gris) is formed. On heating in air, the metal is readily oxidised,
and the product forms scales which are black on the outside (cupric
oxide, CuO), but are red on the side which was in contact with the
metal (cuprous oxide, Cu20). If heated for a long time in air,
cupric oxide is formed.

These two oxides correspond with the cuprous and cupric salts,
in which copper is respectively univalent and bivalent. In solution,
these salts form the cuprous ion, Cu', and the cupric ion Cu",
respectively. The latter is blue ; the former (which readily decom-
poses into the cupric ion and metal : 2Cu! = Cu" + Cu) is colour-
less.

CUPRIC COMPOUNDS, CuX2.

Cupric oxide, CuO. — Cupric oxide, or black oxide of copper, is
formed by the prolonged heating of the metal in air or oxygen, or
by heating cupric nitrate. It is a black solid which is stable up to
its melting point (about 1100°), but then evolves oxygen and leaves
a solution of cuprous oxide, Cu2O, in copper, which forms a solid
solution on cooling. Cupric oxide is readily reduced by hydrogen,
carbon, or organic substances, when heated below redness, and the
metal remains. The oxide dissolves in the borax bead, colouring
it green. If a little tin or stannous chloride is added to the bead, the
cupric oxide is reduced to cuprous oxide, which forms an opaque
red bead. In this way the green copper bead may be distinguished
from that produced by ferrous compounds. Cupric oxide is used
to give a green colour to glass.

When cupric oxide is dissolved in dilute acids, blue solutions of
cupric salts are formed, e.g., CuO + H2SO4 = CuSO4 + H2O.
Concentrated hydrochloric acid gives a yellowish-green solution
of cupric chloride, CuCl2.

EXPT. 319. — Heat a spiral of copper gauze in a large Bunsen flame ;

812 INORGANIC CHEMISTRY CHAP.

a black layer of oxide is formed. Reheat the gauze and place it in a
test-tube containing a few drops of methyl alcohol. The oxide is at
once reduced to bright red copper.

Cuprie sulphate, CuS04. — The commonest cupric salt is the
sulphate, CuSO4, commonly known simply as copper sulphate. This
crystallises from water in large blue triclinic crystals, CuS04,5H2O,
called blue vitriol, or bluestone. It is obtained by dissolving cupric
oxide in dilute sulphuric acid, or copper in hot concentrated sul-
phuric acid (p. 492) : Cu + 2H2S04 = CuSO4 + 2H20 + S02.

According to Cundall, the latter reaction leads first to the formation
of cuprous sulphate, Cu2SO4 ; if the acid liquid is cooled and poured
into water, a red precipitate of copper is formed : Cu2SO4 =
CuSO4 + Cu. Cuprous sulphide, Cu2S, is also formed :

1. 8Cu + 4H2SO4 = 3Cu2SO4 + Cu2S -f 4H2O.

2. 2Cu + 2H2SO4 = Cu2SO4 + 2H2O + SO2.
Secondary reactions then occur :

3. 5Cu2SO4 + 4H2SO4 = Cu2S + 8CuSO4 + 4H2O.

4. Cu2S + 2H2SO4 = CuS + CuSO4 + 2H2O + SO2.

5. CuS + 4H2S04 = CuS04 + 4SO2 + 4H2O.
Equations (1) and (3) give Pickering's equation :

5Cu + 4H2SO4 = Cu2S + 3CuSO4 + 4H2O.

Copper sulphate is prepared on the large scale by the action of
dilute sulphuric acid on copper in the presence of air :

2Cu -f 2H2S04 + 02 - 2CuS04 + 2H20,

or by the " weathering " of copper pyrites, which may first be
roasted : CuS + 2O2 = CuS04. Van Helmont (1644) obtained it
by heating copper with sulphur, and exposing the moistened sulphide
to air : he was thus able to prove that the salt contained copper.
The preparation from copper and oil of vitriol was described by
Glauber in 1648.

Commercial cupric sulphate usually contains ferrous sulphate,
with one hydrated form of which, FeSO4,5H2O, it is isomorphous
and forms mixed crystals. If the solution contains a considerable
amount of copper, the crystals consist of (Cu,Fe)SO4,5H20 ; if the
iron predominates they have the composition (Fe,Cu)S04,7H20.
Similar results are obtained with zinc sulphate. One salt may be
said to " induce " the crystallisation of the other in a particular
form. In order to purify the salt from iron, a concentrated solution
of it is boiled with a little nitric acid. The iron is oxidised to ferric
sulphate, which is not isomorphous with copper sulphate and is
more soluble, hence it is left in solution on crystallisation, and pure
cupric sulphate separates. A solution of the salt containing ferrous
sulphate is used for steeping seeds to prevent " smut," and copper

xxxix COPPER, SILVER, AND GOLD 813

sulphate is employed in calico-printing, in the preparation of pig-
ments (e.g., Scheele's green, CuHAs03), and in electro-deposition.
The salt is insoluble in alcohol ; it is precipitated in small crystals,
CuS04,5H20, when alcohol is added to the aqueous solution.
Several crystalline hydrates of CuSO4 are known (p. 204) ; on
heating the blue pentahydrate crystals to 100° they crumble to a
bluish-white powder of monohydrate, CuSO4,H20. At 220-260°,
this loses most of the combined water, but 0-04 per cent, is retained
even at 360°, and the salt begins to lose sulphur trioxide at higher
temperatures before all the water is expelled. The last molecule
of water of crystallisation of a salt is often retained much more
tenaciously than the others, and for that reason it was called water
of constitution by Graham. The white powder obtained by de-
hydration at 260° is used in the detection of traces of moisture in
alcohol, ether, etc., since it very readily absorbs water and becomes
blue in colour. Anhydrous, or hydrated, copper sulphate readily
absorbs hydrogen chloride, and is decomposed by the aqueous
acid : CuS04 -f 2HC1 = CuCl2 -f H2S04. This reaction may be
applied in separating hydrochloric acid from other gases, such as
sulphur dioxide.

Cupric sulphide, CuS. — Cupric sulphide is a black solid formed
by heating copper powder with excess of flowers of sulphur to a
temperature below 440°, or by precipitating an acid solution of a
cupric salt with sulphuretted hydrogen. In the moist state it is
rapidly oxidised by air, forming a blue solution of the sulphate.
It is slightly soluble in yellow ammonium sulphide, and a red com-
pound, NH4CuS4, may be obtained from the solution. Cupric
sulphide is less stable than cuprous sulphide, and loses sulphur
when strongly heated alone, or in hydrogen : 2CuS = Cu2S + S.

Cupric nitrate, Cu(N03)2. — Copper nitrate is prepared by dissolving
the metal, oxide, .or carbonate in dilute nitric acid, and on evapora-
tion forms blue, deliquescent, prismatic crystals, Cu(N03)2,3H.2O.
At 24-5°, a hexahydrate separates. On heating, the salt loses
water, and also nitric acid, forming a basic salt, Cu(N03)2,3Cu(OH)2.
Copper nitrate possesses powerful oxidising properties : if a few
crystals are moistened and wrapped in tinfoil, sparks are emitted.
The anhydrous salt is obtained as a white powder by the action
of a solution of nitrogen pentoxide in nitric acid on the crystalline
hydrate.

Cuprie halogen compounds. — Cupric chloride, CuCl2, is obtained in the
anhydrous form as a dark brown mass by burning copper in excess
of chlorine, or by heating the hydrate, CuCl2,2H20. It is formed
as a yellow powder by adding concentrated sulphuric acid slowly to
a concentrated solution of cupric chloride. When strongly heated,
it loses chlorine and leaves cuprous chloride (p. 225). A crystalline
hydrate, CuCl2,2H20, is formed in emerald-green crystals by dis-

814 INORGANIC CHEMISTRY CHAP.

solving cupric oxide in concentrated hydrochloric acid and evapo-
rating. In concentrated solutions it is yellowish -green ; on adding
concentrated hydrochloric acid the colour becomes yellow. This
is due to the reversal of the ionisation : CuCl2 r± Cu" + 2C1', the
colour of the undissociated salt being yellow. A very dilute solution
shows the pure blue colour of the cupric ion, Cu" ; the green solu-
tions probably contain a mixture of the blue ion and the yellow
un-ionised salt. Cupric chloride is very deliquescent, and is also
soluble in alcohol. The alcoholic solution burns with a fine green
flame. A green flame is also formed by heating a little cupric
oxide moistened with hydrochloric acid on a platinum wire in a
Bunsen flame, or by heating the oxide in the flame and passing a
little hydrochloric acid gas into the air-hole of the burner. Pure
cupric oxide imparts no colour to the flame, but if moistened with
chloroform, or an organic compound containing chlorine, a green
flame results. This is used as a test for halogens in organic com-
pounds.

An oxychloride of copper, 3CuO,CuCl2,4H2O, is formed as a pale
blue precipitate when caustic potash is added to an excess of cupric
chloride solution. This compound occurs in Atacama, Peru,
Bolivia, etc., in the form of a green sand called atacamite. It is
being formed by the action of sea-water on copper pyrites on the
south coast of Chile. The oxychloride is prepared for use as a
pigment, called Brunswick green, by boiling copper sulphate solution
with a small quantity of bleaching powder.

Cupric bromide, CuBr2, is formed in black crystals by evaporating
a solution of the oxide in hydrobromic acid in a vacuum desiccator over
sulphuric acid. In solution, it shows the same colour changes as the
chloride. Cupric iodide is not known (p. 817).

Cupric hydroxide. — If caustic potash or soda is added to a solution
of a cupric salt, a pale blue gelatinous precipitate, usually regarded
as the hydroxide, Cu(OH)2, is formed, insoluble in excess of alkali.
It appears, however, that the precipitate is, as was stated by
Berthollet (p. Ill), a basic sulphate, CuS04,3Cu(OH)2. If a little
of the copper solution is added to an excess of concentrated alkali,
however, a deep blue colloidal solution of the hydroxide is formed.
If the pale blue hydroxide is boiled with water, it becomes black, a
hydrated oxide of the composition 4CuO,H2O, which is granular
and easily filtered, being formed. On heating to redness, this is
converted into the oxide, CuO.

Cupric carbonates. — Only basic carbonates of copper are
known ; the most important are the minerals chessylite (or
azurite), 2CuC03,Cu(OH)2 (deep blue), and malachite, CuC03,Cu(OH)2
(bright green). The green, deposit (verdigris) formed on copper
exposed to air has the same composition as malachite. If sodium

xxxix COPPER, SILVER, AND GOLD 815

carbonate solution is added to a solution of a cupric salt, carbon
dioxide is evolved and a light blue precipitate of CuC03,Na2CO3J3H2O
is formed : sodium bicarbonate precipitates 5CuO,3CO2,Aq.

Other cupric salts. — Cupric phosphate, Cu3(PO4)2,3H2O, is formed
as a blue, crystalline powder by dissolving the basic carbonate in dilute
phosphoric acid and heating to 70°. Basic phosphates occur as
minerals. The phosphide, Cu3P2, is obtained as a black powder by
boiling phosphorus with copper sulphate solution. When heated in
hydrogen, it forms cuprous phosphide, Cu3P. The black precipitate
formed from copper salts and phosphoretted hydrogen is Cu5P2,H2O.
Copper silicide, Cu2Si, is a grey compound obtained from the elements
in the electric furnace. Copper containing 1-2 per cent, of silicon is
hard, but has a good conductivity for electricity ; it is used for sliding
contacts. Copper orthosilicates, CuH2SiO4, and CuH2SiO4,H2O, occur
as the minerals dioptase, and chrysocolla, respectively.

Copper peroxides, of the formulae Cu2O3 and CuO2,H2O, are obtained
as yellow powders by electrolysing concentrated caustic soda solution
with a copper anode, and by allowing the hydroxide to stand in contact
with hydrogen peroxide for several days, respectively. The com-
pound CuO2,H2O is stable when dry.

I II

CUPROUS COMPOUNDS, CuX, or (Cu2)X2.

Cuprous oxide, Cu20. — Red cuprous oxide, Cu20, is* formed by the
partial reduction of cupric compounds in the presence of alkalies.

EXPT. 320. — Dissolve 69 gm. of pure copper sulphate in 1 litre of
water, adding 1 drop of sulphuric acid. Call this Solution A. Dissolve
in another litre of water 350 gm. of Rochelle salt (sodium potassium
tartrate, NaKC4H4O6,4H2O) and 100 gm. of caustic soda. Call this
Solution B. Mix together 25 c.c. of A and 25 c.c. of B : the resulting
deep blue liquid is called Fehling's solution. Boil this in a porcelain
dish with a solution of glucose (grape sugar). A yellow precipitate of
cuprous oxide, Cu2O, is deposited, which quickly turns bright red.
Filter, wash with boiling water, and alcohol, and dry in a steam-oven.

Cuprous oxide gives a red colour to the borax bead. When
fused with glass it forms the cheaper kind of ruby glass (cf. p. 835).
When treated with dilute sulphuric acid, a solution of cupric sul-

Shate is formed, and metallic copper separates : Cu2O -f- H2SO4 =
u2S04 + H2O ^r Cu + CuSO4 -f H20. Dilute nitric acid dis-
solves the oxide with evolution of oxides of nitrogen, and a solution
of cupric nitrate is formed. Concentrated hydrochloric acid dissolves
cuprous oxide with formation of a colourless solution of cuprous
chloride, Cu2Cl2, or a complex acid, H2CuCl3. The solution rapidly
becomes green on exposure to air, owing to oxidation and formation

816 INORGANIC CHEMISTRY CHAP.

of cupric chloride : 4CuCl + 4HC1 + O2 = 4CuCl2 + 2H2O. The
solution of cuprous chloride in hydrochloric acid is used in gas
analysis for absorption of carbon monoxide. The solution in
ammonia, which is colourless if metallic copper is present, is used to
absorb acetylene.

Cuprous chloride, Cu2Cl2. — By heating copper with mercuric
chloride, Boyle (1664) obtained cuprous chloride, Cu2Cl2, or CuCl, as
a brown, resinous mass, turning green on exposure to air ; he called
it resin of copper. It is formed when copper burns in a limited
supply of chlorine, or hydrochloric acid is passed over heated copper :
2Cu -f- 2HC1 = Cu2Cl2 + H2. Copper does not dissolve in Con-
centrated hydrochloric acid unless air is admitted, when cupric
chloride is formed : 2Cu + 4HC1 + 02 = 2CuCl2 + 2H20. Cuprous
chloride is most easily prepared by dissolving cuprous oxide in
concentrated hydrochloric acid, or by reducing a solution of cupric
chloride in concentrated hydrochloric acid, and pouring the solu-
tion into water. A white precipitate of cuprous chloride is thrown
down.

The reduction of the cupric chloride may be effected by : (a)
boiling with copper turnings until the solution becomes colourless :
CuCl2 -f Cu = Cu2Cl2 ; (6) treating with zinc-dust : 2CuCl2 + H2 =
Cu2Cl2 + 2HC1 ; or (c) passing sulphur dioxide through the solu-
tion : 2CuCl2 + H2SO3 + H20 = Cu2Cl2 + H2S04 + 2HC1.

EXPT. 321. — Dissolve 25 gm. of cupric oxide in 250 c.c. of concen-
trated hydrochloric acid in a flask. Add 50 gm. of copper turnings,
and boil in a fume-cupboard until the solution is colourless. Pour
the solution into a litre of previously boiled distilled water, filter off the
cuprous chloride in a Biichner funnel, wash rapidly with boiling water,
alcohol, and ether. Dry in a vacuum desiccator on a porous plate
over sulphuric acid.

Cuprous chloride is a white powder which crystallises from con-
centrated hydrochloric acid in white tetrahedra. It melts at 434°,
forming a brown, resinous mass on cooling. If exposed to light
when moist it becomes dark coloured (cf. AgCl) ; in moist air it
forms green cupric oxychloride, CuCl2,3CuO,4H20. It dissolves
readily in ammonia, forming a colourless solution of cupro-ammo-
nium chloride, Cu(NH3)Cl,H20, if all traces of oxygen are excluded.
Crystals of this compound are obtained by boiling copper powder
with a solution of ammonium chloride, and cooling. The colourless
solutions in hydrochloric acid and ammonia readily absorb oxygen,
becoming green and blue, respectively, and carbon monoxide,
forming a solution of an unstable compound, 2CuCl,CO,H20.
Acetylene forms a bright red precipitate of cuprous acetylide,
Cu2C2. This is explosive when dry ; when warmed with hydro-
chloric acid, it evolves acetylene : Cu2C2 + 2HC1 = Cu2Cl2 + C2H2.

xxxix COPPER, SILVER, AND GOLD 817

Cuprous iodide, Cul, is precipitated as a white powder on addition
of potassium iodide to a solution of cupric sulphate. Cupric iodide,
if it is first produced, is at once decomposed into cuprous iodide and
free iodine : 2CuSO4 + 4KI = 2CuI -f 2K2S04 + I2. If sulphur
dioxide or ferrous sulphate is previously added, the formation of
iodine is prevented :

2CuS04 4- 2FeS04 + 2KI = 2CuI + Fe2(S04)3 + K2S04.

If the iodine liberated in the first reaction is titrated with thio-
sulphate, the volumetric estimation of copper by this reaction is
possible.

Cuprous sulphate, Cu2S04. — This salt is formed to some extent
when cupric sulphate solution stands in contact with copper :
Cu" + Cu ±^: 2Cu', or Cu2". This is the cause of the inaccuracy
of the ordinary copper coulometer. The pure salt is obtained as
a white powder by heating cuprous oxide with dimethyl sulphate
at 100°, washing with ether, and drying in vacuo. It is at once
decomposed by water, with deposition of copper : Cu2SO4 :^±
CuS04 4- Cu. Cuprous sulphite, Cu2SO3.H2O, is formed as a white
precipitate on passing sulphur dioxide through a solution of cuprous
acetate in acetic acid. Cuprous sulphide, Cu2S, is a black, brittle
mass formed when copper burns in sulphur vapour.

EXPT. 322. — Place a few pieces of roll sulphur on the bottom of a
small flask, and half fill the latter with copper turnings. Heat the
flask : the copper glows with a red light, and a black mass of cuprous
sulphide is formed. Moisten with water and expose to air ; a blue
solution of cupric sulphate is produced.

Cuprous cyanide, CuCN. — If potassium cyanide solution is added
to a solution of cupric sulphate, the yellow cupric cyanide first
precipitated rapidly decomposes with evolution of cyanogen, and
white cuprous cyanide is formed. This dissolves in a solution of
potassium cyanide, forming a colourless solution of potassium
cuprocyanide, KCu(CN)2, which is a salt of a complex anion, since
it ionises as follows: KCu(CN)2 — K* 4- Cu(CN)2'. Only traces
of copper ions from the further ionisation : Cu(CN)2' ±^ Cu" + 2CN',
are formed, and the solution is not precipitated by sulphuretted
hydrogen, since the concentration of copper ions is not sufficient
to exceed the solubility product of the very sparingly soluble cuprous
sulphide.

Potassium thiocyanate gives with a solution of cupric sulphate
to which ferrous sulphate or sulphur dioxide has been added a
white precipitate of cuprous thiocyanate, CuCNS.

Other cuprous compounds. — Cuprous hydride, CuH, is a very unstable
yellow precipitate obtained by reducing a solution of copper sulphate
with a hypophosphite at 70°. It evolves hydrogen with hydrochloric

3 G

818 INORGANIC CHEMISTRY CHAP.

acid : CuH + HC1 = H2 + Cud. Cuprous nitride, Cu3N, is a dark
green powder formed by heating cuprous oxide in ammonia gas.

If copper sulphate solution is added to a solution of sodium stannite,
obtained by adding an excess of caustic soda solution to stannous
chloride, an olive-green precipitate of copper suboxide, Cu4O, is thrown
down. If this is added to dilute sulphuric acid, a colourless solution is
formed. This, after a few seconds, becomes deep purple in colour, and
deposits red metallic copper.

Cuprammonium compounds. — Cupric hydroxide readily dissolves
in ammonia (which precipitates it from a cupric salt), forming a
deep blue solution, known as Schweitzer's reagent. This dissolves
cellulose (filter-paper, cotton-wool), and if the solution is then
squirted into dilute acid, a thread of amorphous cellulose is formed,
which is one variety of artificial silk. The solution may also be
applied to canvas to form a water-tight coating of amorphous
cellulose (Willesden canvas) : some method of preserving cellulose
by impregnation with copper was known to the ancient Egyptians.
The blue ammoniacal solution appears to contain the complex
cuprammonium cation, Cu(NH3)4" : if a solution of cupric sulphate
is precipitated with ammonia, and the precipitate dissolved in excess
of ammonia, a deep blue solution is formed. If a layer of alcohol
is poured carefully over this solution in a cylinder, the latter corked
to prevent evaporation, and the whole allowed to stand, long,
transparent, deep blue rhombic prisms of cuprammonium sulphate,
Cu(NH3)4S04,H2O, are deposited. This salt may be regarded as
CuS04,5H20, in which 4 of the molecules of water of crystallisation
are replaced by molecules of ammonia. Cupric chloride forms
cuprammonium chloride, Cu(NH3)4Cl2,2H2O, which crystallises on
cooling a hot solution of cupric chloride saturated with ammonia
gas. Anhydrous cupric chloride absorbs ammonia gas, forming
the compound CuCL,6NH3, which readily dissociates on heating,
forming CuCl2,4NH3"and CuCl2,2NH3.

SILVER. Ag = 107-04.

Silver.— Silver has been known from the earliest times ; its
association with the moon led to the name Luna, or Diana, given
to it by the alchemists, who represented it by the symbol of the
crescent moon, ([. It is not oxidised by pure air or oxygen, either
in the cold or when heated, and is an example of a noble metal
(silver, gold, platinum). In ordinary air it slowly tarnishes, and
becomes covered with a thin adherent film, which exhibits the
colours yellow, blue, and black, with increasing thickness. This
film is composed of silver sulphide, Ag2S, formed by the decom-
position of hydrogen sulphide present in the air : H2S + 2Ag =

xxxix COPPER, SILVER, AND GOLD 819

The staining of silver spoons used with eggs is also due to the forma-
tion of silver sulphide from the combined sulphur in the albumin of
the egg. The tarnish is readily removed by a dilute solution of potass-
ium cyanide, followed by washing in plenty of water.

Silver occurs frequently in the native state, often in large masses,
in Norway, Peru, and Idaho, occasionally nearly pure, but usually
containing copper and gold. Important ores of silver are the
sulphide, argentite (or silver glance), Ag2S (the commonest ore) ;
chlorargyrite (or horn-silver), AgCl ; pyrargyrite (or ruby -silver],
Ag3SbS3 ; stromeyerite (or silver-copper glance), (Cu,Ag)2S ; stephanite,
Ag5SbS4. Less important are proustite, Ag3AsS3, the bromide,
AgBr, and the iodide, Agl. Traces of silver occur in sea- water
(Proust, 1787).

Metallurgy of silver. — Silver is extracted from its ores by several
processes, the most important being :

(1) alloying with lead, and removing the lead by oxidation

(cupellation) ;

(2) alloying with lead, followed by the separation of silver from

the argentiferous lead by dissolving it in fused zinc (Parkes

(3) amalgamation with mercury, and separation of the mercury

from the silver by distillation ;

(4) dissolving out the silver salts from the ore by a solution of

common salt, sodium thiosulphate, or potassium cyanide,
followed by precipitation (wet processes).

The cupellation process is the most ancient. In it, the silver ore
is smelted with a lead ore, and the resulting alloy of silver and lead
is treated to separate the
silver. The lead obtained from
galena is nearly always argenti-
ferous, and forms an important
source of silver. Formerly
the alloy was treated directly
by melting it on a flat dish
formed of bone ash (Fig. 390).
called a cupel, and a blast of
air driven over the surface of
the metal (Fig. 391). The lead
is oxidised to lead monoxide, or
litharge, PbO, which fuses and
is swept away by a blast of air.
The last portions of litharge are
absorbed by the porous cupel, and a bright button of silver
is left. In the last stage of the process the litharge film becomes so
thin that iridescent colours are seen ; the bright silver surface then

3o2

FIG. 390.— Cupel.

820

INORGANIC CHEMISTRY

CHAP.

" flashes " out. In Germany, the furnace -hearth is formed of
marl, and the cupellation is performed in one operation instead of
the alloy being added in successive quantities, as in the English
process. Alloys containing considerable amounts of lead, such as
the argentiferous lead from galena, are treated to effect a partial
separation before cupellation. This is carried out in two ways,
known as the Pattinson process and the Parkes process.

The Pattinson process (1833). — If fused argentiferous lead is
cooled, a point is reached when pure lead separates out in crystals.
This will occur at a temperature below the freezing point of pure
lead, because of the depression of freezing point by the dissolved
silver (p. 768). The crystals of lead are withdrawn by perforated
iron ladles, and the remaining liquid alloy becomes increasingly

rich in silver
until, if the
process were
carried far
enough, lead
and silver
would begin to
separate out
together at
the eutectic
point. In prac-
tice, seven-
eighths of the
original lead
is removed.
The process is
carried out -in
a row of iron
pots, the lead
separated

being passed on from pot to pot to be remelted, and the liquid alloy
passed in the other direction. The silver gradually accumulates in
the alloy at one end of the series, and desilvered lead at the other.
The rich alloy is then cupelled. In the modification known as
the Luce-Rozan process, only two pots are used, an upper or melting
pot, and a lower or crystallising pot, holding 7 and 21 tons respec-
tively. The lead is deposited in the latter by blowing steam at
50 Ib. pressure through the fused metal, whilst cold water is sprayed
on the surface. When two-thirds of the lead has separated, the
liquid is strained off through a perforated plate. The separated
lead is remelted and the process repeated until the proportion of
silver retained mechanically in the lead crystals is sufficiently small.
The Parkes process (1850). — Molten lead can dissolve only 1-6

FIG. 391. — Cupellation Furnace.

xxxix COPPER, SILVER, AND GOLD 821

per cent, of zinc, and molten zinc can take up only 1-2 per cent, of
lead. Silver, however, is soluble in zinc. If, therefore, zinc is
added to molten lead containing silver, the molten alloy of zinc and
silver floats to the surface, and solidifies on cooling. It is skimmed
off with a perforated ladle, and strongly heated with carbon in a
fireclay retort. Zinc distils off, leaving silver, which is cupelled.
The zinc alloy may also be electrolysed (as anode) in zinc chloride
solution ; zinc is deposited on the cathode, and silver is left. To
remove traces of zinc dissolved in the lead, the latter is heated to
redness and a blast of steam forced through it, when zinc oxide
rises to the surface, leaving the lead. For a ton of lead containing
14 oz. of silver, only 224 Ib. of zinc are required. This process is
superseding the Pattinson method.

Any gold present is also removed by the zinc. The desilvered lead
contains only 0-0004 per cent, of silver, whilst that obtained by the
Pattinson process contains 0-00 1-0 '002 per cent.

Amalgamation and wet processes. — The amalgamation process has
been used in Mexico, where fuel is scarce ; it was invented by
Bartolomeo de Medina in 1557. The ores, containing metallic
silver, silver chloride and sulphide, and a large quantity of rock, are
finely crushed in stamping mills worked by mules, and the fine
mud is mixed with a little salt. The mass is then well trodden
by mules on a paved floor, or patio. Mercury is then added, together
with a little roasted pyrites, containing cupric and ferric sulphates,
and the treading is continued for fifteen to forty-five days. Copper
chlorides are probably first produced from the roasted pyrites
and salt, and these decompose the silver sulphide with formation of
the chloride : CuCl2 + Ag2S = 2AgCl + CuS ; or 2CuCl + Ag2S =
Cu2S -f- 2AgCl. The silver chloride then dissolves in the brine,
and is reduced by the finely-divided mercury : AgCl -f Hg =
Ag -f- HgCl. The silver forms an amalgam with the excess of
mercury. (About 1 per cent, of sodium is now added to the mer-
cury, to prevent the latter forming a fine powder, which would be
lost in washing.) The amalgam is separated by washing, the calomel
being lost, the excess of mercury is pressed out from the amalgam
in canvas bags, and the residue distilled in iron retorts to recover
the mercury. This process has, since 1904, been gradually replaced
by the cyanide process (see below).

In the wet processes the ore is roasted, either alone, when soluble
silver sulphate is formed and can be lixiviated, or with salt, when
silver chloride is produced, which is extracted with salt solution,
or a solution of sodium thiosulphate. From these solutions, the
silver is precipitated by sodium sulphide as silver sulphide. In the
modern cyanide process, the unroasted ore, finely ground in ball
mills, is leached with a 0-7 per cent, solution of sodium cyanide,

822

INORGANIC CHEMISTRY

the slime being well agitated by a stream of air. Soluble sodium
argento-cyanide, NaAg(CN)2, is formed, the sodium sulphide also
produced, which would tend to stop the reaction, being oxidised
to thiosulphate and sulphur by the current of air : Ag2S -f- 4NaCN
^± 2NaAg(CN)2 + Na2S. The silver is precipitated from the
solution by scrap zinc.

Refining of silver. — Silver is refined by cupellation, or by the

Moebius electrolytic process (1884). The electrolyte consists of silver
nitrate solution with about 1 per cent, of free
nitric acid ; the cathode is a plate of pure silver
and the anode a block of the silver to be refined.
\l ^1 Silver is deposited, copper dissolves, and the gold

FIG 399 —cupel present in the anode alloy deposits as a slime. The
copper must not accumulate in the solution

beyond 4-5 per cent. The gold slime is collected in a canvas

bag round the anode.

Silver alloys. — Commercial silver is alloyed with copper, because

the pure metal is too soft for coinage or jewelry work. The pro-
portion of silver in 1,000 parts of alloy is called the fineness. British

silver coin since the time of

Edward I. has had a fineness

of 950 ; in France, Germany,

and Austria the silver coinage

has a fineness of 900. The

metal still retains the pure

white colour of silver.

The composition of the alloy

used by the Mint is ascer-
tained yearly in a public trial,

known as the trial of the pyx,

conducted by competent as-

sayers appointed by the

Goldsmiths' Company, who also

carry out trials with gold coin.

The assay is made by heating

a weighed portion of the alloy

with a little pure lead on a

bone-ash cupel (Fig. 392) in a

muffle furnace (Fig. 393). This

is a furnace in which a fireclay oven containing the cupels is strongly

heated on the outside, the mouth of the muffle being only loosely

closed, so as to admit air. The copper is oxidised, and the oxide

dissolves in the lead oxide, which is easily fusible, and is absorbed

by the cupel. The trial of the cupel is described by the Latin

Geber.

FIG. 393.— Muffle Furnace for Cupellation.

xxxix COPPER.. SILVER, AND GOLD 823

Silver goods are often treated by heating in air ; the copper in the
alloy oxidises ; the oxide is removed by dilute acid, leaving a surface of
pure silver. Test -portions must therefore be taken from the mass of
the metal.

Pure silver. — In order to obtain pure silver from its alloy with
copper, the latter is dissolved in dilute nitric acid, when copper
nitrate, Cu(N03)2, and silver nitrate, AgN03, are formed. The
solution is evaporated to drive off some of the excess of acid, and
diluted with water. Hydrochloric acid is added in slight excess.
A curdy white precipitate of silver chloride, AgCl, is produced. This
is filtered off and washed with hot water till free from acid. To obtain
silver from the chloride it is treated in one of several ways.

(a) The chloride is fused in a crucible with sodium carbonate, when a
button of pure silver is formed : 4AgCl -f 2^X00, = 4Ag + 4NaCl +
2C02 + 02.

(b) The silver chloride is boiled with a solution of caustic potash and
grape-sugar : the oxide is first formed as a dark-brown powder, which
is then converted into a grey powder of metallic silver, together with a
dark-brown solution containing the oxidation products of the sugar :
2AgCl + 2NaOH = Ag2O + 2NaCl + H2O ; Ag2O = 2Ag + O.
The silver is then well washed with boiling distilled water.

(c) Dilute sulphuric acid is poured over the silver chloride, and a stick
of pure zinc placed in the mixture. The chloride is reduced by the
nascent1 hydrogen, forming a grey mass of silver powder, which is
washed and dried on a water-bath : AgCl + H = Ag + HC1. The
silver from (b) or (c) may be fused in a crucible with sodium carbonate
to form a button. (If silver is fused in a glazed porcelain crucible, the
latter becomes yellow, owing to the formation of silver silicate.) Stas
distilled silver in a lime crucible with the oxy -hydrogen blowpipe.
Richards (1905) showed that Stas's silver probably contained a little
occluded oxygen, which may be removed by heating in a vacuous tube
at 400°.

Fused silver dissolves oxygen, which is liberated as soon as the
metal begins to solidify. Ten gm. of silver at 1020° dissolve 20-5 c.c.
of oxygen (at S.T.P.). The solid crust is violently disturbed, and
the metal " spits." This phenomenon, which is easily observed
with the metal fused on a cupel, is shown only by pure silver, and
is a good test of the completion of cupellation.

Properties of silver. — Silver melts at 962° in the absence of air,
and at 956° in air. It boils at 1955°, forming a blue vapour, the
density of which corresponds with the formula Ag. The density
of the solid is 10-5 ; it contracts on fusion, like ice, and the solid
floats on the molten metal. Silver is very malleable and ductile ;

824 INORGANIC CHEMISTRY CHAP.

it can be beaten into leaves 0-00025 mm. thick, which become some-
what transparent on heating. Very thin films deposited on glass
(cf. below) also transmit blue light.

Silver is attacked by boiling concentrated sulphuric acid, or
cold dilute nitric acid, but resists the action of alkalies, even when
the latter are fused. Silver crucibles are therefore used in the
laboratory for fusion with caustic alkalies, but may be replaced by
those of pure nickel, although the latter is slightly attacked.

Silver deposited on glass by reduction is used in the manufacture
of mirrors.

EXPT. 323. — Clean a test-tube with boiling nitric acid, wash well with
water, and prepare in it a solution of silver nitrate. Add dilute ammonia
drop by drop until the precipitate of silver hydroxide is almost redis-
solved. Then add caustic potash and a solution of Rochelle salt
(potassium sodium tartrate — this acts as the reducing agent). Place
the tube in a beaker of water and heat the latter to boiling. A mirror of
silver is deposited on the tube.

Colloidal silver. — A colloidal solution of silver may be prepared
by Bredig's method of striking an electric arc between silver wires
under water. The metal is volatilised, and condensed in the water
in the form of very small particles, which remain in colloidal sus-
pension. Colloidal solutions are also formed by reduction.

EXPT. 324.; — To 200 c.c. of a 10 per cent, solution of silver nitrate add
quickly a solution formed by mixing 200 c.c. of a 30 per cent, solution of
ferrous sulphate and 200 c.c. of a 40 per cent, solution of sodium citrate,
the mixture being carefully neutralised with soda before addition to the
silver solution. A lilac precipitate of silver is formed, which is filtered
off. and washed with a 5-10 per cent, solution of ammonium nitrate. It
is then soluble in pure water, forming a red, transparent solution. If
silver nitrate, ferrous sulphate, or magnesium sulphate is added to the
red solution the colloidal silver is coagulated, and is then no longer
soluble in water, whereas silver thrown down from the solution by
ammonium nitrate redissolves on washing. The former salts are
adsorbed by the silver precipitate. If the precipitates are dried, they
form blue solids.

Carey Lea (1889) considered these substances to be allotropic
modifications of silver ; it is now recognised that they are ordinary
silver in the colloidal condition (see colloidal gold, p. 834). By
heating silver nitrate with an alkaline solution of sodium protalbate
or lysalbate, Paal (1902) prepared a yellow solution of colloidal
silver. If this is dialysed, and evaporated on a water-bath, a
brownish -black powder, containing as much as 93 per cent, of
silver, and soluble in water, is formed. This is called collargol.

XXXIX

COPPER, SILVER, AND GOLD

825

The colloidal modifications of silver do not conduct electricity ;
on heating, they all give ordinary silver. Traces of the substances
present in solution are always adsorbed by the colloid1, which does
not seem to exist in a perfectly pure state.

Electroplating with silver.— The electro-deposition of silver takes
place in a very direct manner, free from secondary reactions, and is
applied in the silver coulometer for the measurement of current
strength. The International ampere is defined in terms of such
an instrument (p. 282), as the current which deposits 0-001118 gm.
of silver per second from a specified silver solution.

The apparatus in ordinary use (Fig. 394) for measurement of currents
consists of a platinum dish or crucible, which is carefully weighed, and
contains a solution of silver nitrate (300 gm. per litre). The dish is
placed on a brass plate on the base of the stand, which is connected with
the negative terminal. The anode is a rod of pure silver, suspended by a
clamp in the solution. A small glass cup is suspended under the anode,
to retain detached pieces of the latter. The
crystalline deposit of silver adheres to the dish ;
it is washed with water and alcohol, the dish dried
in an air -oven, and weighed. If the experi-
ment lasts for t seconds, and m mgm. of silver
are deposited, the mean current strength is
m/(t X 0-001118) ampere.

FIG. 394.— Silver
Coulometer.

Copper articles are electroplated with silver
by cleaning their surfaces and suspending
them in a solution of silver cyanide in
excess of potassium cyanide, the anode being a plate of pure silver.

The solution contains the complex compound potassium argento-
cyanide, KAg(CN)2, which ionises in a similar manner to the copper
compound : KAg(CN)2 z± K* -f Ag(CN)2'. The anion is very slightly
dissociated : Ag(CN)2' ^ Ag' + 2CN', and the silver ions are de-
posited on the cathode, as a coherent film of metal instead of the crystal-
line metal, which is formed from silver nitrate solution. The disso-
ciation of the complex ion proceeds as silver ions are withdrawn from
the solution. The cyanide ions are discharged on the silver anode,
forming silver cyanide, which dissolves in the solution. The net
result is the transfer of silver from the anode to the cathode.

This process was invented by Wright, of Birmingham, in 1840, and a
patent was taken out by the firm of Elkington, which still produces electro-
plated goods. Previous to the use of electro -plating, copper goods were
plated with silver by laying a strip of silver on a bar of clean copper,
heating, and rolling the bar to the required thickness. This is known
as Sheffield plate, and the layer of silver is much thicker than in the case
of electro-plated goods.

826 INORGANIC CHEMISTRY CHAP.

Compounds of silver. — Unlike copper and gold, silver forms only
one series of compounds, in which the element is univalent. It
does not form basic salts, a tendency which is prominent in the
case of copper. The silver salts are largely ionised in solution, the
silver ion, Ag!, being split off. Salts of gold do not ionise in this
simple way. The test for the silver ion is the formation of a white
curdy precipitate of silver chloride, AgCl, when a solution of a
chloride is added. This precipitate is insoluble in dilute nitric
acid, but is readily soluble in ammonia, potassium cyanide, or
sodium thiosulphate. In all cases complex compounds, which
give only a few silver ions in solution, are formed when the silver
chloride dissolves. The concentration of silver ions is not sufficient
to exceed the solubility product of silver chloride.

Silver nitrate, AgN03. — The most important salt of silver is the
nitrate, AgN03, the preparation of which is described by the Latin
Geber : " Dissolve silver in aqua fortis (aqua dissolutiva) ; boil in
a phial with a long neck, not stopped, until one-third has been con-
sumed (evaporated), and finally set in a cold place. You obtain small
fusible stones, transparent as crystal." The salt forms large trans-
parent rhombic plates, melting at 209°. The fused salt may be cast
into sticks, and is then used as a cautery under the name of lunar
caustic. The alchemists also called it lapis infernalis. It is readily
decomposed by organic matter, such as paper, cork, or the skin,
metallic silver being deposited. The silver is deep black in colour,
so that a solution of silver nitrate is used as an indelible ink for mark-
ing linen. The same black stain is formed on the skin ; it can be
removed from the articles by a dilute solution of potassium cyanide.
Silver nitrate is poisonous, but is given internally in small doses
in nervous diseases. The salt is soluble in alcohol.

Silver nitrate decomposes when strongly heated ; oxygen and
nitrogen dioxide are evolved, and silver remains. The decom-
position point is much higher than that of copper nitrate, so that
copper nitrate may be separated from silver nitrate by heating,
adding water, and filtering from the copper oxide. Copper may
also be separated by adding a little caustic soda to a portion of the
solution, filtering off and washing the silver oxide, Ag2O, and boiling
it with the rest of the solution. Copper oxide is precipitated, and
the silver oxide goes into solution as nitrate.

Solid silver nitrate absorbs ammonia gas, with evolution of heat,
and formation of a compound AgN03,3NH3. If ammonia is added
to a solution of the nitrate until the oxide first precipitated is dis-
solved, and the liquid evaporated, crystals of a compound
AgNO3,2NH3 separate. This is the nitrate of a complex cation,
Ag(NH3)2'. Double salts, e.g., AgN03,2NaN03 and AgNO3,KN03,
are known.

Silver nitrite, AgNO2, is formed as a crystalline precipitate when

xxxix ^COPPER, SILVER, AND GOLD 827

solutions of silver nitrate and potassium nitrite are mixed. It
decomposes on heating, evolving oxides of nitrogen.

Silver oxide, Ag20. — Finely-divided silver, when heated to 300°
in oxygen under pressure, forms a brown oxide, Ag2O. If caustic
soda is added to a solution of silver nitrate, a brown precipitate
of silver oxide, Ag2O, is thrown down. This may be dried at
60-80°, and is then black. The hydroxide, AgOH, is said to be
precipitated from alcoholic silver nitrate by alcoholic potash at
— 30°. The oxide may also be prepared by boiling the chloride
with caustic soda solution. It is very slightly soluble in water
(3-10~7 gm. mol. per litre), forming a solution alkaline to litmus,
and the moist solid readily attracts carbon dioxide from the air,
producing silver carbonate, Ag2CO3. The latter is precipitated as
a light yellow powder when a carbonate is added to a solution of
silver nitrate. With excess of potassium carbonate, a double
carbonate, KAgCO3, is formed as a white precipitate.

Silver oxide gives off oxygen at 300°. It is used as an oxidising
agent in organic chemistry, and for giving a yellow colour to glass,
a yellow silicate, Ag2SiO3, being formed. When the oxide is dis-
solved in ammonia, and the solution exposed to the air, a black pre-
cipitate of the nitride, Ag3N, is deposited. This is very explosive
when dry, and is called fulminating silver.

A peroxide, Ag204, is deposited, mixed with silver nitrate, on the
anode in the electrolysis of silver nitrate with platinum electrodes.
It evolves oxygen on heating.

Halogen compounds of silver. — Silver fluoride, AgF, is the only
halogen compound of silver appreciably soluble in water. Hydro-
fluoric acid does not act on the metal, but dissolves the oxide.
On evaporation in a vacuum, crystals of AgF,H2O are deposited,
which cannot be completely freed from water by heating. By
evaporating the solution in the air, very deliquescent crystals of
AgF,2H2O are formed. The fused salt, which contains metallic
silver (4AgF + 2H2O = 4Ag -f- 4HF -f 02), is an elastic black mass,
easily cut with scissors. Silver fluoride, under the name of tacky ol,
has been proposed for sterilising water.

Silver chloride, AgCl, occurs native as horn-silver, described by
Conrad Gesner (1565) as argentum cornu, and by Matthesius (1585)
as " glass-ore, transparent like horn in a lantern." It is readily
prepared as a curdy white precipitate by adding hydrochloric acid,
or a chloride, to a solution of silver nitrate ; on heating to 460° it
fuses to a dark-yellow liquid, which solidifies on cooling to a soft,
colourless, tough mass. It was therefore called luna cornea by
Oswald Croll (1608). Silver chloride volatilises at a white heat,
giving the vapour density corresponding with AgCl. The fused
chloride, according to Stas, is quite insoluble in cold water, but the
curdy precipitate is slightly soluble (p. 359). The latter becomes

828 INORGANIC CHEMISTRY ^ CHAP.

powdery on standing in the liquid for a time. Silver chloride dis-
solves slightly in dilute nitric acid on standing ; it dissolves in 200
parts of concentrated hydrochloric acid, is fairly easily soluble in
sodium chloride, and readily in ammonia or sodium thiosul-
phate solution. The solution in ammonia contains complex ions :
Ag(NH3)2Cl :=± Ag(NH3)2" -f Cl'. Solid silver chloride absorbs ammo-
nia, forming AgCl,3NH3 and 2AgCl,3NH3 (p. 548). The thiosul-
phate solutions contain a very stable silver sodium thiosulphate,
AgNaS2O3, which separates in crystals on evaporation. It possesses
a strong sweet taste. If a solution of silver nitrate is added to
one of sodium thiosulphate, and the liquid boiled, a black precipitate
of silver sulphide is produced by the decomposition of the silver
thiosulphate, which is transiently formed as a white precipitate :

(H ,

SO + = SO/ + Ag2S.

(OH \OH

The insolubility of silver chloride provides a means for the
estimation of silver (or of chlorides). The precipitate may be heated
until it begins to fuse, and weighed, but as it tends to pass into a
colloidal solution on washing, it is more convenient to adopt the
volumetric method. No indicator is necessary if more than traces
are present, as the curdy precipitate settles after the bottle con-
taining the liquid has been violently shaken, and the silver nitrate
solution (N/10 = 16-01 gm. of AgNO3 per litre) is added till a drop
produces no 'further turbidity in the settled solution. Titration
is carried out in a stoppered bottle covered with a roll of black paper,
to prevent discoloration of the precipitate by light (p. 830). For the
estimation of smaller amounts, a little potassium chromate is added
to the neutral chloride solution before titration with silver nitrate ;
when all the chloride is precipitated, red silver chromate, Ag2Cr04,
begins to be formed, giving a permanent brownish-red colour to the
suspension. Another method is to add a little iron alum to the
solution of the chloride acidified with nitric acid and titrate with
ammonium thiocyanate. When the precipitation of the white
curdy silver thiocyanate, AgCNS, is complete, ferric thiocyanate is
formed which gives a red colour to the solution.

Silver bromide, AgBr, forms a pale yellow precipitate, insoluble
(like the chloride) in dilute nitric acid, and only sparingly soluble
in dilute ammonia. Silver iodide, Agl, is produced as a light yellow
precipitate, quite insoluble in dilute nitric acid, and only very
sparingly soluble in ammonia (which changes its colour to white).

Silver powder dissolves in aqueous hydriodic acid with evolution of
hydrogen ; on cooling, colourless crystals of AgI,HI separate, which
rapidly decompose. Agl is dimorphous ; the bromide and iodide melt at
426° and 556°, respectively. Silver iodide contracts on heating from

xxxix COPPER, SILVER, AND GOLD 829

— 10° to 70° (Fizeau, 1876). Silver bromide does not absorb ammonia
gas ; it dissolves in liquid ammonia, and crystals of AgBr,3NH3 separate,
decomposing at 4° into 2AgBr,3NH3. The iodide forms 2AgI,NH3 with
ammonia gas, and AgI,NH3 with liquid ammonia.

If chlorine is passed into water containing an excess of silver
oxide in suspension, silver chloride and free hypochlorous acid are
first produced. (These are the only products if silver oxide is
added to excess of chlorine water.) The hypochlorous acid reacts
with the excess of silver oxide, forming a solution of silver hypo-
chlorite, AgCIO ; the solution then does not smell of HC1O, but is
still an active bleaching agent :

Ag2O (solid) ±1; Ag20 (dissd.) -f H20 ±^ 2AgOH
AgOH + C12 = AgCl + HC10.
AgOH + HC10 = AgCIO + H2O.

In presence of silver oxide, the hypochlorite is fairly stable, but if
the suspended solid is allowed to settle, the supernatant liquid
rapidly deposits white silver chloride. The liquid loses its bleaching
properties and now contains silver chlorate, AgClO3, which may be
crystallised out and dried in air at 150° (Stas) : 3AgClO =
2AgCl -f- AgClO3. By reducing the chlorate in solution with sul-
phurous acid, silver chloride is formed : AgClO3 -f- 3S02 + 3H20 =
AgCl + 3H2S04.

Silver sulphate, Ag2S04. — This salt is sparingly soluble in water,
and is formed by boiling silver with concentrated sulphuric acid, or
by precipitating a solution of the nitrate with sulphuric acid.
It dissolves readily in dilute or concentrated sulphuric acid, or in
dilute nitric acid. The acid sulphate, AgHS04, is formed in light
yellow crystals when the sulphate is dissolved in less than three
parts of sulphuric acid. Silver sulphide, Ag2S, is formed when
silver is heated with sulphur, or silver nitrate is precipitated with
sulphuretted hydrogen.

The disulphide, Ag2S2, is formed by mixing solutions of sulphur in
carbon disulphide and of silver nitrate in benzonitrile. Silver sulphite,
Ag2SO3, is formed by precipitation ; on heating to 100°, it forms the
dithionate : 2Ag2SO3 = Ag2S2O6 -f 2Ag.

Silver phosphates. — Silver orthophosphate, Ag3P04, is formed as
a pale yellow precipitate when a solution of sodium phosphate is
added to one of silver nitrate. The reaction is usually represented
by the equation :

3AgN03 + Na2HPO4 = Ag3P04 + 2NaN03 + HN03,

but as the precipitate is readily soluble in nitric acid, only two-
thirds of this amount of silver is precipitated :

6AgN03 + 3Na2HP04 = 2Ag3PO4 + 6NaN03 + H3P04.

830 INORGANIC CHEMISTRY CHAP.

The acid phosphate, Ag2HPO4, is deposited in white crystals from a
solution of the phosphate in phosphoric acid. The metaphosphate,
AgPO3, and pyrophosphate, Ag4P2O7, are gelatinous and crystalline
white precipitates, respectively, formed by adding silver nitrate
to the corresponding sodium salts. Silver arsenite, Ag3As03, and
silver arsenate, Ag3As04, are canary-yellow and brick-red, respec-
tively, and are formed by precipitation. The former dissolves in
ammonia, and if the solution is boiled, silver is deposited.

Silver phosphide, AgP2, is formed by the union of the elements at
400°. The acetylide, Ag2C2, is formed as an explosive white precipitate
by passing acetylene into an ammoniacal solution of silver nitrate.

Photography. — The blackening of silver chloride on exposure to
light was observed by Boyle, who explained it as due to the action
of air. Scheele (1777) showed that if the blackened substance is
digested with ammonia, unchanged silver chloride is dissolved and
a residue of silver remains. He also noticed that the violet rays
act most strongly on the chloride, whilst the red and orange rays
have practically no action. Silver salts may be rendered sensitive
to these rays by adding to them certain dyes (aurin, erythrosin,
cyanin) which absorb light of these wave-lengths. These sub-
stances are called photo-sensitisers.

The first to turn the sensitive silver salts to account in making light
pictures, or photographs, was Thomas Wedgwood (1802). In 1839
Daguerre allowed iodine vapour to act on a polished silver surface,
which was exposed in the camera, and an invisible image was produced.
The treated plate was exposed to mercury vapour, which condensed only
on the portions which had been acted on by light, leaving the unaltered
iodide in the shadows. The iodide was removed by a solution of sodium
thiosulphate, as suggested by Herschel, and the picture thus fixed,
or rendered non-sensitive to light. Archer (1851) used a transparent
film of collodion (p. 570) spread on glass, and impregnated with zinc or
cadmium bromide or iodide. This was treated before use by immersion
in a solution of silver nitrate, when the halide was deposited. The
plate was exposed in the camera whilst still wet, and then developed
(Talbot, 1841) by immersion in a solution of a reducing agent such as
ferrous sulphate, or pyrogallic acid, which converted the altered halide
into black metallic silver. The unaltered halide was then dissolved out by
potassium cyanide or sodium thiosulphate, and a negative produced, in
which the light and shade in the picture are reversed. To obtain a
positive, the plate is laid on a piece of paper coated with silver chloride,
and then exposed to light. The print is fixed in the same way as the
plate ; it is not developed, as the chloride can be directly changed
sufficiently in colour to give the requisite shades.

In the modern dry-plate process, an emulsion of silver bromide is

xxxix COPPER, SILVER, AND GOLD 831

prepared by adding ammoniacal silver nitrate to a solution of gela-
tin in hot water containing potassium bromide. The emulsion,
after " ripening " at 45° for some time, whereby the halide grains,
at first of diameter 0-001 mm., grow to 0-003 mm., is poured in a
thin film on a glass plate or celluloid film. The soluble salts are
washed out after setting, and the film dried, all operations being
carried out in the dark or in orange light-! After exposure (which
may only be a fraction of a second), the plate is developed with
pyrogallol, hydroquinone, or metol (paramethylaminophenol) in
presence of alkali. These substances are oxidised, and the silver
bromide which has been changed by light is reduced to metallic
silver, e.g.,

C6H4(OH)2 + 2AgBr = C6H402 + 2Ag + 2HBr

Hydroquinone Quinone

To prevent over-vigorous development, when some of the un-
changed bromide is reduced and leads to " fogging " of the plate, a
retarder, consisting of potassium bromide, is added to the developer.
After washing, the plate is fixed in sodium thiosulphate. The print,
or positive, is made on silver chloride paper, coated with albumin,
which is toned after exposure by immersion in a solution of gold
chloride (brown tone), or platinic chloride (grey tone), a little of the
silver being dissolved and replaced by the nobler metal. It is then
fixed in thiosulphate.

A print may also be made on silver bromide paper (velox, or gaslight
paper), which is exposed in the same way as a plate, and developed.
The gelatin in the plate and the albumin on the paper act as
sensitisers to the silver salt, accelerating the action of light upon it.

The exact mechanism of these photochemical changes is still
obscure. According to one theory, a sub-halide, e.g., Ag2Br, is
formed by loss of halogen, which is absorbed by the sensitiser. But
hydrobromic acid is never detected in the gelatin, the whole of the
bromine passing into the developer. Pure dry silver chloride does
not blacken on exposure to light, and the ordinary salt always takes
up oxygen as it loses chlorine, so that it has been suggested that an
oxy-chloride is produced, Ag2ClO. Recent work points, however,
to a purely physical explanation (Joly, 1905). Halides of silver
on exposure to light emit electrons, and the photo-sensitiveness
is in . the proportion of the order of the photo-electric activity :
AgBr>AgCl>AgI. Cathode rays (free electrons), and JC-rays
(which produce free electrons from matter) also produce photographic
effects. The molecules of halide which have lost an electron are
supposeelifco be those capable of being developed. Scheele's original
experiments, however, prove conclusively that chemical reactions
occur when the action of light is prolonged.

832 INORGANIC CHEMISTRY CHAP.

GOLD. Au = 195-6.

Gold. — Gold, by reason of its occurrence in the free state, and of its
beautiful colour and brilliance, was probably the first metal known
to man. Gold ornaments are found in neolithic remains. Because
of its supposed perfection, the metal was associated by the alchemists
with the sun, was called Sol, and represented by the symbol Q,
or €J. The alchemists considered that the other metals, if suitably
purified, or " cleansed," could be transmuted into gold.

A typical description of the art of making gold, as understood by the
alchemists, is the following from Philalethes, " Secrets Revealed in
Chymistry " (1669) : " That thou mayest have this knot well unfolded,
attend diligently. Let there be taken of our Fiery Dragon, which hides
the magical Chalybs in his own belly, 4 parts ; of our Magnet 9 parts ;
mix them well together with a torrid Vulcan, or great fire, in the form of
a mineral water, upon which there will swim a scum, which is to be cast
away ; remove the shell and then the kernel, purge it the third time with
Fire and Salt, which will easily be done if Saturn shall have beheld
himself in the looking glass of Mars. Thence is made the Chameleon, or
our Chaos in which all Arcana's lies hid virtually but not actually."
The author adds : " None ever wrote so clearly." It is instructive
to compare this with the " Sceptical Chymist " of Robert Boyle, published
in the same year : " For Quicksilver, with several Metals, will compose
an Amalgama ; and with divers Menstruums, it seems to be brought into
the form of a Liquor ; and with Aqua-Fortis, it may be turned either
into a white or into a Red Precipitate ; with Oyl of Sulphur, into a pale
Yellow one ; with Sulphur it will compose a red, and Volatile Cinabar :
With some Saline Bodies, it will ascend in the form of White Salt,
Dissoluble in Water ; with Regulus of Antimony, and Silver, it may be
Sublimed into Chrystal : and . . . yet out of all these Substances,
it may be again Obtained, and Reduced to its Pristine Form." There
is nothing in this description which cannot be followed by the modern
chemist.

Gold occurs usually in the native condition, alloyed with a certain
amount of silver, and sometimes copper and traces of platinum.
Some tellurium compounds of gold occur in small amounts (p. 531),
and traces of gold are found in pyrites and other ores, and in sea-
water. Gold is recovered from burnt pyrites, but a sea-weed
which collects gold instead of iodine is yet unknown. The native
gold occurs in quartz veins or reefs intersecting metamorphic rocks
f the chlorite type, such as occur in Wales, where gold extraction is
o/ried on to a small extent. The most important European locali-
car^here gold is found are Russia, Hungary, and Germany. Hun-
ties \vcrold may contain tellurium, which must be separated if the
garian 3 to be used for dental purposes. Gold occurs all over
metal ihe Russian mines of the Urals, discovered in 1819, supplied
Asia. 'Ihe metal until the accidental discovery of gold in California
most of t,The richest fields are in Africa, especially the Transvaal
in 1849.

xxxix COPPER, SILVER, AND GOLD 833

Rand (which supplied 8,237,700 oz. in 1911, and gives the highest
production in the world) and in Australia. In North America the
fields extend from Mexico to Klondike, the latter field being opened
in 1896.

Metallurgy of gold. — Native gold occurs either as nuggets of vary-
ing size (one of 1841b. weight was found at Ballarat) in the rock, or as
grains in the alluvial sand. From the latter it is extracted by wash-
ing away the lighter sand in agitating cradles or sluices, or breaking
up the auriferous gravel by powerful jets of water, under 100-300ft.
head. The rock is crushed in batteries of stamping mills, and mer-
cury is added to the resulting slime to amalgamate with the gold.
The gold amalgam is retained by amalgamated copper plates. The
residual slime (" tailing ") is treated by the cyanide process (q-v.).
The amalgam is scraped off the plates, distilled in iron retorts, and the
residue cupelled (p. 819).

To extract gold from auriferous pyrites, obtained from the rock
as so-called " concentrates," they are treated by Plattner's chlorine
process. The roasted pyrites are moistened with water in tubs with
false-bottoms, and impregnated with chlorine gas. After 24 hours
the soluble gold trichloride, AuCl3, is leached out with water,
and the gold precipitated by ferrous sulphate. The reaction is
one of simultaneous oxidation and reduction : AuCl3 + 3FeSO4 =
Au + FeCl? + Fe2(S04)3, or? more simply : Au'" -f 3Fe" =
Au -f 3Fe"". Bromine water has also been used.

Gold is now extracted on a large scale from the tailings from
stamp-mills or directly from the finely stamped ore by the cyanide
process of MacArthur and Forrest. The slimes are agitated in large
tanks with a solution of cyanide containing 0-3 per cent, of KCN,
in whicli the gold dissolves. After settling, or filter-pressing, the
clear liquor is reduced by adding metallic zinc (of which metal the
packages for the export of the cyanide are made). The precipitate
is fused with lead and the alloy, containing 10 per cent, of gold, is
cupelled. The reactions in the cyanide process are somewhat
complicated ; they occur in presence of atmospheric oxygen, and
hydrogen peroxide is formed (p. 342) by autoxidation :

2Au + 4KCN + 2H20 + 02 = 2KAu(CN)2 + 2KOH + H202

Potassium
aurocyanide

2Au + 4KCN + H202 = 2KAu(CN)2 -f 2KOH.

The reduction process is : 2KAu(CN)2 + Zn = K2Zn(CN)4 + 2Au.
In this way quartz containing only half an ounce of gold per ton
can be profitably treated.

Gold refining. — The gold bullion is then refined. If it contains
copper, this is removed by an oxidising fusion with borax and nitre.
The silver and gold are separated by granulating the alloy, and boil-

3 H

834 INORGANIC CHEMISTRY CHAP.

ing with concentrated sulphuric acid, which extracts the silver as
sulphate, leaving the gold (Scheele, 1753). The alloy must not
contain more than 33 per cent, of gold, otherwise the silver is not
dissolved. If it contains more gold than this the alloy is melted with
silver until it contains about one-quarter its weight of gold. This
operation of separation is termed quartation. In the new electrolytic
process of Wohlwill (1910), the bullion is made the anode in a solution
of 2 «5-6 per cent, of gold chloride, containing 2-5 per cent, of hydro-
chloric acid, and an alternating current is superposed on the direct
electrolysing current. In the Rose process (1910) the zinc precipitates
are fused, and air or oxygen is blown through, when the baser
metals oxidise in succession and pass into a borax-silica flux.

Standard gold. — Pure gold is too soft for use as ornaments or for
coinage, and it is alloyed with copper, or silver, or both. The copper
makes the colour redder (English gold coin), silver imparts a pale
colour (Australian gold coin). The fineness is expressed either in
parts per 1000, or in carats, pure gold being 24 carat fine, and five
standard alloys of 22, 18, 15, 12, and 9 carat, i.e., parts of gold in
24 of alloy, are legalised. The 22 carat English gold coin has a
fineness of 916-67. German, Italian, and American coinage has a
fineness of 900, i.e., 21-6 carat. The presence of 1 part of bismuth
in 1920 parts of gold renders the metal brittle.

Gold plating is carried out in the same way as silver plating, by
electro-deposition from solutions of gold cyanide in potassium
cyanide, the requisite amounts of silver and copper salts being added.
These metals are deposited as an alloy with the gold if a suitable
voltage is used.

Properties of gold. — Gold is a bright yellow metal, which crystal-
lises (like most metals) in the regular system ; it has a high density
(19*32), and is a good conductor of heat and electricity. It melts at
1064°, expanding on fusion, and forms a bluish-green liquid (cf.
copper, p. 810). It volatilises appreciably 100° above its melting-
point, and boils at about 2500°. It is the most ductile metal, as was
stated by Pliny, and can be beaten into leaves 0-0005 mm. thick.
The deposits on gold lace are only 0-000002 mm. thick. By treating
gold leaf with a solution of potassium cyanide, Faraday obtained
films 0-0001 mm. thick, which transmit green light. On heating
gold-leaf the metal crystallises and minute gaps are formed, which
transmit red light, as does ruby-glass (q.v.). Gold is not attacked
by oxygen or ariy single acid except selenic, but it dissolves in solu-
tions of chlorine, bromine, or iodine, and therefore in aqua regia.
Fused alkalies and nitrates, and a solution of ferric bromide,
also attack it.

Colloidal gold is formed by Bredig's process (p. 824), or by reducing
solutions of gold chloride with phosphorus, ferrous sulphate,

xxxix COPPER, SILVER, AND GOLD 835

hydrazine, formaldehyde, etc. The different solutions have different
colours according to the sizes of the colloidal particles. Those with
larger particles are blue ; with decreasing size the colour passes
through a fine ruby-red to yellow, and approaches that of gold
chloride, containing gold atoms, in an unbroken chain of perfect
continuity. This indicates that there is no fundamental difference
between colloidal and true solutions. By precipitating a mixture of '
gold, stannous, and stannic chlorides, with alkali, a purple powder,
called purple of Cassius (discovered by Andreas Cassius, and described
by his son in 1685), used for making ruby glass, is thrown down. It
appears to be a colloidal form of tin oxide with adsorbed colloidal
gold (Moissan, 1905). When glass is fused with purple of
Cassius and annealed, it assumes a fine ruby colour, due
to the presence of ultra-microscopic particles of gold. Gold differs
from copper and silver in the extreme ease with which its com-
pounds are reduced to the metal.

Compounds of gold. — If gold is dissolved in aqua regia it forms a
bright yellow solution, which on evaporation deposits deliquescent
yellow crystals of chlorauric acid, HAuCl4,3H20, commonly known
as " gold chloride." The solution is reduced to metallic gold by
hydrogen gas or exposure to light. If gold is dissolved in chlorine
water, the solution evaporated, and the residue heated to 150°, a
brown, crystalline mass of auric chloride, AuCl3, is left, soluble in
water, alcohol, and ether. On heating AuCl3 at 175° it gives off
chlorine and leaves a yellow powder of aurous chloride, AuCl, which at
higher temperatures decomposes into chlorine and gold. AuCl is de-
composed by water : 3 AuCl = AuCl3 + 2Au. Chlorauric acid, when
evaporated with a solution of potassium chloride, gives red crystals
of the potassium salt 2KAuCl4,H2O. On heating, these form the
chloraurite, KAuCl2. If AuCl is treated with cold dilute potash, a
violet powder of aurous oxide, AuO, is left. By precipitating chlor-
auric acid with potash, auric hydroxide, Au(OH)3, is obtained. This
is a weak base, and also dissolves in caustic potash, the solution
depositing potassium aurate, KAu02,3H20, on evaporation in
vacuo. The hydroxide when gently heated forms auric oxide,
Au203, which at a higher temperature readily decomposes into
oxygen and the metal, Auric bromide, AuBr3, is formed by dissolving
gold in bromine water ; if gold is heated with iodine at 50-114°,
aurous iodide, Aul, is formed. On precipitating gold chloride with
potassium iodide, a green precipitate of auric iodide, AuI3, is first
formed, which quickly decomposes into the aurous compound and
iodine (cf. copper).

The sulphides, Au2S and AuS, are formed when H2S is passed into
solutions of potassium aurocyanide (afterwards acidified), and a cold
neutral solution of AuCl3, respectively : 8AuCl3 -f 9H2S + 4H2O ~

3 H2

836 INORGANIC CHEMISTRY CHAP.

8AuS + 24HC1 + H2SO4. Au2S3 is not formed in the latter reaction,
but is produced when anhydrous lithium aurichloride is treated with
H2S at— 10°.

By fusing gold with sodium sulphide and sulphur, it forms a substance
soluble in water, and by evaporation in a vacuum colourless crystals of
sodium aurosulphide, NaAuS,4H2O, are deposited. Stahl suggested that
this was the method used by Moses in reducing the Golden Calf to a
potable form for the consumption of the Israelites. From a solution of
auric chloride in sodium thiosulphate, colourless crystals of Fordos and
Gelis' salt, Na3Au(S2O3)2,2H2O, separate. This substance is not
reduced by ferrous sulphate.

Fulminating gold, or auro-diamine, AuHN-NH2, is prepared by
digesting precipitated auric hydroxide with ammonia ; it is a dirty
olive-green powder which when dry explodes with great violence
when heated or struck. It is decomposed by hydrochloric acid :
AuN2H3 + 5HC1 = AuCl3 + 2NH4C1. By the action of ammonia
on aurous oxide, NAu3-NH3 (sesquiaurammine) is formed, which
on boiling with water forms gold nitride, Au3N.

An important compound of gold is potassium aurocyanide,
KAu(CN)2, used in electro-plating. This is produced by dissolving
fulminating gold in boiling potassium cyanide solution. On cooling
colourless lustrous crystals separate. From the solution hydrochloric
acid precipitates yellow aurous cyanide, AuCN. Auric cyanide,
Au(CN)3, is not known, but potassium auricyanide, KAu(CN)4, also used
in electro-gilding, is obtained in colourless crystals by mixing hot
concentrated solutions of gold trichloride and potassium cyanide.

A delicate test for gold is the purple precipitate formed by pouring
the solution into concentrated stannous chloride solution (1 part of
gold in 100 million parts of water can be detected).

The atomic weight of gold has been found by the analysis of
potassium auribromide K2AuBr4, and other salts. The accepted
value is 195-6 (H = 1).

EXERCISES ON CHAPTER XXXIX

1. Give a general description of the properties of the metals : copper,
silver, and gold, with special reference to their position in the Periodic
System.

2. In what forms does copper occur, and how is the metal obtained ?

3. What alloys of copper, silver, and gold are used, and for what
purposes ?

4. How are the following prepared from metallic copper : cupric
oxide, cuprous chloride, cupric sulphate ? What are their properties ?

5. How are cuprous oxide, cuprous sulphate, and cuprammonium
sulphate prepared ?

6. What is the action of (a) potassium iodide, (6) potassium cyanide,

xxxix COPPER, SILVER, AND GOLD 837

7. Give a brief account of the occurrence and metallurgy of silver.
How would you obtain a specimen of pure silver from an alloy with
copper ?

8. Describe the preparation and properties of (a) colloidal silver,
(b) silver oxides, (c) fulminating gold, (d) gold ruby-glass, (e) silver
sulphate, (/) gold trichloride.

9. How have the atomic weights of silver and gold been determined ?
Why is the atomic weight of silver of great importance in connection
with those of other elements ?

10. What chemical reactions occur in photography ? How are
photographic plates prepared ?

11. How are silver- and gold-plating carried out electrolytically ?
What is electrotyping ?

CHAPTER XL

THE ALKALINE-EARTH METALS

Metals of the alkaline earths. — The elements of Group II in the

Periodic Table are all metals. They are divided into two sub-groups,
the odd series and the even series : —

(a) Even series : the metals of (6) Odd series : beryllium, mag-
the alkaline earths : cal- nesium, zinc, cadmium, and

cium, strontium, barium, mercury,

and radium.

With the possible exception of mercury, all these metals are
bivalent in all their compounds : RX2. The mercurous salts, such
as calomel, HgCl, in which the metal seems to be univalent, probably
have the doubled formulae Hg2X2, in which the group — Hg — Hg — ,
made up of two bivalent mercury atoms, is also bivalent. All these
metals form basic oxides, RO, and (except mercury) hydroxides,
R(OH)2. There is a regular gradation in the solubility of these
hydroxides in series (a) ; those of series (6) are practically insoluble
in water. The same holds for the chlorides, RC12, of sub-group (a) : —
Grams dissolved by 100 gm. of water :

Ca(OH)2 0-29 at 10° CaCl2 74-5 at 20°

Sr(OH)2 0-92 „ Sr012 53-9 „

Ba(OH)2 3-9 „ BaCl2 35-7 „

The older chemists gave the name earth to all non-metallic sub-
stances insoluble in water and unchanged by fire. Lime and
magnesia were found to have an alkaline reaction, and were called
alkaline earths, the name being afterwards applied to baryta (Scheele,
1774), and strontia (Hope, 1792). The metals themselves were
isolated by Davy (1807) by electrolysis, following a suggestion by
Lavoisier that, like other " bases," they were oxides of metals.
Magnesium is now usually placed in sub-group (b).

The metals of the alkaline earths are all silver-white, oxidise in the
air, and decompose water, though less vigorously than the alkali-
metals sodium, potassium, etc. They form, in addition to the basic
oxides, RO, true peroxides, R02, in which the metal is still bi-

OH. XL THE ALKALINE. EARTH METALS 839

valent : R/^ | . They unite directly with hydrogen and with nitro-
gen, forming hydrides, RH2, and nitrides, R3N2, respectively. The

Fia. 395.— Forms of Calcite.

compounds give distinctive colours when heated on platinum wire,
moistened with hydrochloric acid, in the Bunsen flame : calcium,
orange-red ; stron-
tium, crimson ; bar-
ium, apple-green.

CALCIUM, Ca=39-75.
Limestone. -- The
most abundant min-
eral of calcium is the
carbonate, CaC03,
which is dimorphous,
crystallising in var-
ious forms of the
hexagonal system as
caldte (Fig. 395), and

in the rhombic Sys- FIG. 396.— Crystals of Aragonite.

840

INORGANIC CHEMISTRY

CHAP.

tern as aragonite (Fig. 396). The former is the commoner form ;
besides occurring in minerals, it forms the chief constituent of egg-
shells, bones (together with calcium phosphate), oyster-shells, and
coral, all of which effervesce with acids. In the massive form it
occurs as marble, limestones of various kinds, calc-spar (a very pure
transparent variety of which is Iceland spar), and chalk. Chalk
consists of the shells of minute marine organisms. In combination
with magnesium carbonate, it forms dolomite, MgC03,CaCO3, of
which (as well as limestone) whole mountain-chains are composed.

If carbon dioxide is passed through cold lime-water, the amorphous
flocculent precipitate which first appears soon becomes crystalline,
and has the form of
calcite. If the lime-
water is hot, crystals
of aragonite are pro-
duced. Calcite is the
stable form at the
ordinary temperature ;
at 470° it is converted
into aragonite. A
third form, /x-CaC03,
is said to be precipi-
tated at 60°. The
solubility of calcium
carbonate in water
containing carbonic
acid has already been
described (p. 205).
By adding a solution
of KHCO3 to a cooled
solution of CaCl2, a
white precipitate of
Ca(HC03)2 is formed.

FIG. 397. — Derbyshire

Quicklime, CaO. — If calcium carbonate is heated to dull redness

(550°), it begins to decompose, evolving carbon dioxide, and leaving
calcium oxide, or quicklime, CaO. In a closed vessel, the decomposi-
tion stops at a certain pressure of the carbon dioxide, known as the
dissociation pressure, which has a definite value at each temperature ;
the system is then in equilibrium : CaCO3 ^± CaO + C02.

The dissociation pressure at various temperatures is given below ;

it increases rapidly with the temperature :

t° 700° 750° 800° 850° 900°

Pco2mm. Hg. 50 99 195 370 700

If the carbon dioxide is swept away by a current of air, dissociation

goes on till the reaction is practically complete. This is applied in

THE ALKALINE -EARTH METALS

841

the manufacture of quicklime from limestone or marble. The process
is known as lime-burning.

In some districts, e.g., in High Peak, Derbyshire, the limekiln
is filled with blocks of the limestone, and an arch of lumps of the
stone is built over the fire below, the fuel being kept separate from
the stone (Fig. 397). The burning goes on for thirty-six to forty-
eight hours, when the kiln is allowed to cool, and the lumps of quick-
lime (which is then nearly pure — " Buxton lime " contains 98 per
cent, of CaO) are removed. This process is wasteful in fuel, and on the

Continent continuous
limekilns (Fig. 398). are
used. The broken stone
is mixed with about 20
per cent, of its weight
of coke or coal, and is
fed intermittently into
a shaft kiln through a
cup-and-cone arrange-
ment, a, b. The coke
burns, and the C02
produced from the
CaC03, mixed with
nitrogen, passes out
through d. The lime
works its way gradually
down the kiln, and is
withdrawn through
apertures, e at the base.
It contains the fuel
ashes, and is therefpre
less pure than that
made in intermittent
kilns.

Pure calcium oxide is prepared by heating Iceland spar with the
blowpipe in a platinum crucible, with free access of air, until a little
of the white opaque residue, after cooling and addition of water, no
longer effervesces with acid. It is a white, amorphous mass, sp. gr.
3-3, which melts at about 1900° and can be boiled in the electric
furnace, the vapour condensing in cubic crystals. Lime resists the
temperature of the oxy-hydrogen blowpipe without more than sinter-
ing, and is therefore used in making furnaces for fusing platinum.
These consist (Fig. 399) of two blocks of lime, hollowed out, in the
lower one of which the metal is placed, whilst the blowpipe is
introduced through a hole in the upper block. The electric furnace
used by Moissan (Fig. 400) was also constructed of lumps of quick-
lime.

Lime Kiln.

842

INORGANIC CHEMISTRY

CHAP.

Slaked lime, Ca(OH)2. — If quicklime is moistened with water,
much heat is evolved, and clouds of steam are given off. ( Gunpowder
may be kindled by strewing a little over the mass.) The lime com-

Fiti. 398. — Continuous Lime
Kiln.

FIG. 399. — Oxyhydrogen Blow Pipe
Furnace with Lime Crucible.

bines with the water, cracks, and finally, after addition of sufficient
water, crumbles down to a fine, dry, white powder. This is calcium
hydroxide, Ca(OH)2, known as slaked lime. If mixed with an excess
of water a paste, called milk of lime, is formed ; if shaken with water
it dissolves sparingly, producing lime-water (2-2 gm. of CaO per litre
at 10°). The solubility, as Dalton found, decreases with rise of

temperature (p. 99). Calcium
hydroxide is also formed as a
white precipitate by adding
caustic soda to a concentrated
solution of calcium chloride :
CaCl, + 2NaOH = Ca(OH)2 -'

2NaCl. With saturated solu-
tions the mixture becomes
solid (" the chemical miracle "
of Francisco Lana, 1686). Six-
sided crystals of calcium
hydroxide are deposited by
evaporating lime-water in a vacuum over sulphuric acid. Slaked
lime, when heated to dull redness, loses water, and is converted
into quicklime.

FIG. 400. — Moissan's Electric Furnace.

XL THE ALKALINE -EARTH METALS 843

Quicklime, when exposed to the air, slowly absorbs moisture and
carbon dioxide, crumbling to a powder which consists of a mixture of
hydroxide and carbonate. Pure lime does not absorb perfectly
dry carbon dioxide. Lime-water on exposure to air becomes
covered with a crust of calcium carbonate. If this is broken it falls
to the bottom, and another appears. In this way the whole of the
lime is precipitated.

Lime is used chiefly in the preparation of mortar, for building
purposes, this consisting of a thick paste of slaked lime with three to
four times as much sand as quicklime originally taken. Lime made
from magnesian limestone slakes slowly and is called " poor lime,"
distinguished from " fat lime," which slakes easily. The hardening
of mortar consists in the evaporation of the moisture, or its absorp-
tion by the bricks, and the slow conversion of the hydroxide into
calcium carbonate by atmospheric carbon dioxide ; slight combina-
tion between the lime and the silica of the sand also occurs.
Modern mortar usually contains ground cinders and rubbish ;
soluble salts from these form an efflorescence on the bricks, consisting
chiefly of sodium sulphate.

Calcium peroxide is formed as a hydrate, Ca02,8H2O, by precipitat-
ing lime-water with H202. From very concentrated solutions at
0°, or in all cases above 40°, anhydrous Ca02 is precipitated. Calcium
peroxide is manufactured for use as an antiseptic by compressing
slaked lime and Na2O2, and washing with ice-water. Much free
lime is present in it. It is not formed directly from CaO and 02
(cf. Ba02).

Cement. — If limestone containing more than 5 per cent, of clay is
burnt, the resulting lime forms a mortar which hardens under water,
and is therefore called hydraulic mortar. The old Roman mortar
was of this type, and many buildings constructed with it are still
standing firm. In 1796 J. Parker prepared a similar Roman cement
by heating clay and limestone. Portland cement is made by burning
a mixture of limestone and clay, either mixed with coal as in lime-
burning, or by feeding the mixture into the top of a revolving
tubular furnace inclined at an angle, into the lower part of which a
blast of air charged with coal-dust, which forms an intense flame,
is injected. The cement-clinker so produced is ground to powder,
and packed in air-tight barrels.

The constitution of cement, and the mechanism of setting, have
been variously explained. Cement clinker appears to contain the
following compounds :

tricalcium silicate, 3CaO,Si02

tricalcium aluminate, 3CaO,Al2O3;

calcium orthosilicate, 2CaO,Si02

pentacalcium trialuminate, 5CaO,3Al2O3.

844 INORGANIC CHEMISTRY CHAP.

A certain amount of free lime, CaO, is also present. According to
other investigators, tricalcium aluminate is a solid solution of lime
in pentacalcium trialuminate. On addition of water, the calcium
aluminates are first hydrated, then the silicates take up water.
During this process free lime is separated as calcium hydroxide. Le
Chatelier regarded the final compounds as 2CaSi03,5H2O, and
4CaO,Al203,12H20, unstable supersaturated solutions of the basic
silicates being initially formed, which rapidly crystallise in the form
of a mass of interlacing needles of the basic silicate. Michaelis,
however, considered that the compound 2CaSi03,5H20 is produced
in the first instance as a colloidal jelly, the gradual dehydration of
which leads to the hardening of the cement. The formation of col-
loidal material in the early stages of the setting has been confirmed.

A mixture of cement and broken bricks or gravel forms concrete ;
reinforced concrete is concrete cast over a steel framework.

Calcium chloride, CaCl2. — If limestone or marble is dissolved in
hydrochloric acid, a solution of calcium chloride, CaCl2, is formed.
This usually contains ferric chloride as an impurity, and is yellow.
A little chlorine water is added to oxidise any ferrous iron, then the
solution is filtered, and milk of lime added until the liquid is slightly
alkaline. On boiling, ferric hydroxide is precipitated ; if the
filtered liquid is neutralised with pure hydrochloric acid and evapo-
rated to a syrupy consistency, colourless very deliquescent crystals
of the hexahydrate, CaCl2,6H2O, m.-pt, 30° separate. These
dissolve in water with considerable lowering of temperature ; the
eutectic point is — 55°. On heating the crystals to 200°, water is
evolved, and a white, porous mass of the dihydrate, CaCl2,2H20,
remains, which is used for preparing solutions for refrigerators.
If heated strongly, a porous mass of the anhydrous salt is formed,
which is used in drying gases, etc. This fuses at 780°, and forms a
hard crystalline mass on cooling. The dihydrate and the anhydrous
salt evolve heat when dissolved in water. Calcium chloride dissolves
readily in alcohol. Anhydrous calcium chloride absorbs ammonia
gas, forming the unstable compound CaCl2,8NH3.

If a solution of 120 parts of CaCl2 in 100 parts of water is cooled to
18-38°, a tetrahydrate, CaCl2,4H2O, separates, which exists in two
forms, a and /3. At 45-3°, the stable a form gives CaCl2,2H2O ;
at 177-5°, CaCl2,H2O separates from the solution ; and at 260°,
anhydrous CaCl2 (Roozeboom, 1889).

Large quantities of calcium chloride are formed as a by-product
of the Ammonia-Soda process (p. 782) ; a solution of it is used in
refrigerating plants (" brine "), since it freezes only at a low tem-
perature ; and also, on account of the deliquescent character of the
salt, for preventing dust on roads.

THE ALKALINE -EARTH METALS

845

Homberg (1693) observed that freshly-fused calcium chloride is
phosphorescent; Baldwin (1674) had noticed the same property with
calcium nitrate. Perfectly pure salts are not phosphorescent ; the pro-
perty is due to the presence of traces of heavy metals, such as bismuth.

Calcium fluoride, CaF2, or fluor-spar (m.-pt. 1330°) (p. 415) is nearly
insoluble; the bromide, CaBr2 (m.-pt. 765°), and iodide, CaI2 (m.-pt.
740°), are similar to the chloride.

Metallic calcium. — Metallic calcium is now prepared on a technical
scale by the electrolysis of a mixture of 100 parts of calcium
chloride and 16-5 parts of fluor-
spar, fused at 660° in a bath
formed of blocks of carbon.
The cathode is an iron rod,
which touches the surface of
the fused chloride (Fig. 401).
The cathode is slowly screwed
up as the calcium accumulates,
and the metal is drawn out into
the form of an irregular rod,
20-30 cm. in diameter, pro-
tected from oxidation by a
layer of chloride. The metal
(sp. gr. 1-55) melts at 810°, and
readily sublimes in a vacuum
below this temperature. It is
malleable, burns brightly in
oxygen, combines with sulphur,
chlorine, nitrogen, etc., and
reduces nearly all metallic
oxides on heating. It reduces
sodium chloride at 800°.

WATER

FIG. 401 . — Manufacture of Calcium by
Electrolysis.

Calcium is used in freeing abso-
lute alcohol from the last traces
of water. The liquid is digested with calcium turnings, when a somewhat
violent reaction occurs. A little phosphorus pentoxide is added to
the clear liquid, to combine with traces of ammonia (formed from the
nitride, Ca3N2), and the alcohol is distilled.

If calcium is heated in a tube connected with a nearly evacuated
vessel, it absorbs the last traces of air, forming CaO and CasN2, and a
very high vacuum is produced. Heated calcium is used in separating
argon from nitrogen (p. 600).

By heating calcium and calcium chloride in a steel cylinder to 1000°,
red crystals of the subchloride, CaCl, are formed. CaF and Cal are
also known.

846

INORGANIC CHEMISTRY

CHAP.

By passing hydrogen and nitrogen over heated calcium, the
hydride (" hydrolith," p. 183), CaH2, and nitride, Ca3N2, respectively
are formed. The hydride, which is formed at 400-500° with incan-
descence, is colourless ; the nitride is brownish -yellow ; both are
crystalline. On passing steam over the nitride, ammonia is pro-
duced : Ca3N2 -f 6H2O = 3Ca(OH)2 -f- 2NH3. Ammonia gas is
absorbed by calcium with formation of Ca(NH3)4 and evolution of
heat. This ignites in air ; in absence of air it forms Ca(NH2)2.

Calcium sulphate, CaS04. — Calcium sulphate, CaS04, occurs as
anhydrite along with limestone or rock-salt, or more commonly as the
di-hydrate gypsum, CaS04,2H2O, which forms transparent crystals
called selenite (Fig. 402, often twinned), or occurs in crystalline
masses, either fibrous (satin spar) or opaque (alabaster). Anhydrous
calcium sulphate exists in two forms ; (a) natural anhydrite and the
substance formed by dehydrating gypsum at a red heat, both practi-
cally insoluble ; (b) a soluble form, " setting " with water, produced

by dehydrating gypsum at 60-70°
in a vacuum over P205. Gypsum
can easily be reduced to an extremely
fine powder, and the solubility in-
creases with the fineness of the
grains. This is a general result,
and is due to surface-tension forces,
which are more pronounced with
small particles. The solubility of
gypsum increases with rise of tem-
perature to 40°, and then diminishes
at higher temperatures.

If gypsum is heated to 120-130°
it loses water and forms plaster of
Paris, the hemihydrate, 2CaS04,H20, which when mixed with water
evolves heat, and quickly solidifies, expanding slightly ; it is there-
fore used for making plaster casts in moulds. If the surface is
painted with a solution of paraffin wax in petrol, the wax fills the
pores, and an ivory-like surface is produced. Plaster of Paris,
if heated at 140°, begins to lose water ; the whole of the water is
rapidly expelled at 200° ; the residue of anhydrous CaS04 rapidly
takes up water, but if the heating has been more intense the residue
hydrates only very slowly, and is said to be dead-burnt. By heating
to 400°, slight decomposition into CaO and S03 occurs and the
German plaster called Estrich-gips, which sets slowly, and produces
a smooth, hard surface, used for floors, walls, etc., is formed.

Precipitated gypsum is formed by adding sulphuric acid to a
solution of calcium chloride. It is used under the name of pearl-
hardening for adulterating (" filling ") glazed paper. Barium
sulphate is used for a similar purpose, giving a very heavy paper.

FIG. 402.— Gypsum Crystals.

XL THE ALKALINE -EARTH METALS 847

The double salts, CaS04,K2SO4,H20 (syngenite), CaSO4,Na2SO4
(glauberite), and CaS04,2Na2,SO4,2H2O, are known. Calcium sul-
phate dissolves in a concentrated solution of ammonium sulphate,
forming CaSO4,(NH4)2S04,H2O. Strontium and barium sulphates
are insoluble.

Calcium sulphite, CaS03,2H20, is formed as a white precipitate
by passing sulphur dioxide through lime-water, or adding a solution
of a sulphite to one of calcium chloride. It dissolves in aqueous
sulphurous acid, forming a solution known as calcium bisulphite,
Ca(HS03)2 ; this is prepared by passing sulphur dioxide in excess,
through milk of lime ; it is used in sterilising beer casks, and in the
manufacture of wood-pulp.

Wood consists of cellulose and lignin, the latter soluble in boiling
bisulphite solution. The cellulose is left, and is used for making paper.
The pulp is bleached by chlorine, the excess being removed by sodium
thiosulphate (p. 369). The paper is glazed by adding aluminium
sulphate to the pulp, together with rosin soap, and gypsum as " filling."
Insoluble aluminium resinate is formed '

which, on hot-rolling, becomes glossy.
The paper then ceases to absorb ink.

Calcium sulphide, CaS, is formed as
alkali-waste in the Leblanc process, or
by heating gypsum with charcoal :
CaS04 + 2C =088 + 200,. It is in-
soluble in water, but dissolves when Furnace,

sulphuretted hydrogen is passed through

the suspension, forming the hydrosulphide, Ca(SH)2, which crystallises
as Ca(SH)2,6H2O. The polysulphides, CaS2 to CaS5, or possibly CaS7,
appear to be contained in the reddish -yellow solution of sulphur in
boiling milk of lime (thion hudor, p. 481). The crystals which
separate from concentrated solutions are CaS4,3Ca(OH)2,9H20. The
thiosulphate, CaS203,6H20, is formed by blowing air through a suspen-
sion of the sulphide, or by boiling the sulphite and sulphur with
water. If the solution is precipitated with sodium carbonate, sodium
thiosulphate is formed : CaS2O3 -f Na2CO3 = CaCO3 + Na2S203.
This can be crystallised by evaporation of the filtered solution.

Calcium carbide, CaC2.-- Calcium carbide, CaC2, was obtained by
Wohler (1862) on heating carbon with an alloy of calcium and zinc.
It is now manufactured on a large scale by Moissan's process. A
mixture of 2 parts of coke and 3 parts of quicklime is heated to a
very high temperature (2000°) in a closed electric furnace. The
latter (Fig. 403) may consist of a rectangular tank of fireclay,
divided into three compartments, lined with gas-carbon and
having a graphite block in the base forming one electrode.

848 INORGANIC CHEMISTRY CHAP.

The other electrode consists of three vertical blocks of carbon,
one in each compartment, suspended from chains and gradually
lowered into the furnace as they become consumed. Arcs are
struck between the base-plate and these electrodes, and at the
high temperature reaction occurs, with the formation of fused
carbide, which is tapped off, cooled, and broken into pieces in a jaw-
crusher. The reaction : CaO + 30 = CaC2 + CO, begins at 1620°.
The commercial product is a greyish-black stony mass ; pure calcium
carbide, formed by heating calcium hydride in acetylene, consists of
colourless transparent crystals. Calcium carbide is decomposed
by water, with production of acetylene (p. 678) : CaC2 + 2H20 =
Ca(OH)2 -f- C2H2 ; 1 kgm. of commercial carbide usually gives about
300 litres of gas. Commercial calcium carbide when heated in a
stream of nitrogen, reacts with the formation of a mixture of
calcium cyanamide, and graphite : CaC2 -f- N2 = CaCN2 -f C. About
10 per cent, of calcium chloride or fluoride is usually added as a
flux. The product is used as a fertiliser in agriculture. If it is
heated with water under pressure in autoclaves, with a little soda,
steam being passed in, ammonia gas is produced : CaCN2 -f 3H20 =
CaC03 + 2NH3. This is Frank and Caro's method for the fixation
of atmospheric nitrogen (p. 544).

Calcium carbide is an energetic reducing agent. A mixture of
powdered carbide with ferric oxide and ferric chloride burns violently
when ignited with a taper, and fused metallic iron is produced.

Calcium nitrate, Ca(N03)2. — This salt is present in the soil and
serves as a plant food (p. 696). It is manufactured on a large scale
by neutralising dilute nitric acid made by the arc process (p. 574)
with limestone, and evaporating. It is also produced by passing
oxides of nitrogen into milk of lime, or a suspension of calcium car-
bonate in water, until the nitrite in the mixture of nitrite and nitrate,
first-produced, is decomposed, the oxides of nitrogen produced
passing to more milk of lime (p. 576). The salt forms very deli-
quescent monoclinic crystals, Ca(N03)2,4H20, soluble in alcohol. It
is usually fused and cast into blocks for export. The fused salt is
phosphorescent after exposure to light, and is sometimes called
Baldwin's phosphorus, from its discoverer (1674).

Calcium phosphates. — The mineral phosphates have already been
described (p. 608). Pure calcium orthophosphate, Ca3(P04)2, is
formed as a white, flocculent precipitate on adding ordinary sodium
phosphate to a solution of calcium chloride made alkaline with
ammonia : SCaCL + 2Na2HP04 + 2NH4OH = Ca3(P04)2 +
4NaCl + 2NH4C1 + 2H20 ; or 3Ca" + 2HP04" + 2OH'
Ca3(PO4)2 + H20. The precipitate is nearly insoluble in water, but
is slowly decomposed on boiling into an insoluble basic and a soluble
acid salt. It dissolves in water containing many salts, or dissolved
carbon dioxide (which dissolves the calcium phosphate in the soil

XL THE ALKALINE -EARTH METALS 849

and renders it capable of absorption by the roots of plants.) If a
solution of calcium chloride is mixed with one of ordinary sodium
phosphate, a white precipitate of calcium hydrogen phosphate,
CaHPO4,2H20, is formed. By dissolving either of the preceding
salts in aqueous phosphoric acid, crystals of tetra-hydrogen calcium
phosphate, CaH4(P04)2, are formed on spontaneous evaporation.
They are decomposed by water : CaH4(PO4)2 = CaHPO4 + H3PO4.
A mixture of CaH4(P04)2, and gypsum, known as superphosphate of
lime (Fourcroy and Vauquelin, 1795), is prepared for use as a fertiliser,
by macerating ground calcium phosphate in the form of bone-ash,

Shosphorites, etc., with two-thirds of its weight of sulphuric acid :
a3(P04)2 + 2H2S04 + 4H20 = CaH4(P04)2 + 2CaS04,2H20. A
mixture of chamber and Glover tower acids is used, and the phosphate
is first dried and crushed. The reaction is carried out in a mixer,
consisting of a horizontal cast-iron cylinder with revolving blades
inside. The mixture issues in a nearly fluid state, and drops into
pits or dens, which are half-filled, and then closed. The reaction
takes place with rise of temperature, and gases (C02, SiF4, HF, and
HC1) escape through a vent to absorption towers. After a day or
two, the superphosphate formed is removed by picks or mechanical
elevators, powdered in a crusher, and carefully dried by hot air in
long brickwork chambers. The total production of superphosphate
is about 10 million tons per annum. On heating, the salt
CaH4(P04)2 decomposes, with formation of insoluble calcium pyro-
and meta-phosphates: 4CaH4(P04)2 = 3CaH2P207 + Ca(P03)2 +5IJ20.

Calcium oxalate, CaC204. — This salt is formed as a white precipitate
insoluble in acetic acid but soluble in dilute hydrochloric acid, wh^n
ammonium oxalate solution is added to a solution of a calcium sal
preferably after adding ammonium chloride and ammonia,
gentle ignition it gives the carbonate : CaC204 = CaC03 + C
which on heating to redness leaves the oxide, CaO. These reactions
are used in the gravimetric estimation of calcium ; in the volumetric
method the precipitate of oxalate is washed and dissolved in dilute
hydrochloric acid. Sulphuric acid is added, and the oxalate
titrated with standard permanganate : 2KMn\34 -f- 5CaC204 +
8H2S04 = 2MnS04 + K2S04 + 5CaS04 + 10C02 + 8H20. Calcium
oxalate occurs in small crystals (raphides) in some plants.

Glass. — The arts of making, working, and colouring glass appear to
have been known to the Egyptians about 2000 B.C. From Egypt
they spread to Rome, Constantinople, and Venice. An independent
glass industry was established in Germany in the Middle Ages, and
was introduced later to France and England.

Glass consists of a supercooled, very viscous, liquid solution of
silicates ; if a long glass tube is supported at each end it bends
permanently owing to its slight plasticity, but breaks with sudden
shocks. Cobbler's wax shows similar properties. Common glass

3 i

850 INORGANIC CHEMISTRY CHAP.

contains calcium and sodium silicates, and has approximately the
composition Na20,CaO,6SiO2. Sodium silidate is soluble, but glass
is practically insoluble in water, although boiling water removes
some sodium silicate from it. Glass is made by fusing together silica
(sand, crushed quartz, flints), calcium carbonate (limestone, marble,
chalk), and either soda-ash (Na2CO3) or a mixture of salt-cake
(Na2S04) and carbon : 2NaaS04 + C = 2Na2O + C02 + 2SOa, in
fireclay pots or tanks at about 1200°, and allowing the impurities to
settle. On cooling to about 700°, the liquid becomes plastic, and
can be blown or rolled. This gives ordinary soda-glass.

Bohemian, or potash-glass, contains potassium instead of sodium,
and has a higher melting-point and greater resistance to reagents ;
for these reasons it is better adapted to making chemical apparatus,
Flint-glass is soda or potash glass with lime partly replaced by
lead oxide : litharge (PbO) is used in its manufacture. It has a high
refractive index, and is used for optical purposes, but is very soft.
Special varieties of glass (Jena glasses) invented by Schott and
Abbe of Jena (who published their formulae many years ago),
contain boric, arsenic, and phosphoric anhydrides in place of some
of the silica, and also potassium, zinc, aluminium, and barium
oxides. They are made for various optical and chemical purposes.
If good glass is heated to its softening point for a long time or
inferior glass for a short time, some of the constituents crystallise, and
the glass becomes opaque (devitrification). All varieties of glass
require annealing before use : the objects are heated for a time and
allowed to cool slowly. Toughened glass is obtained by cooling in
oil. Coloured glasses are made by adding various metallic oxides
to the fused glass ; in the case of gold the colour only develops after
reheating the glass for some time to increase the size of the colloidal
particles which appear to be present :

Ruby : gold, or cuprous oxide. Blue : cobalt oxide.

Green : chromium oxide, or Opaque milky glass : tin oxide

cupric oxide. or calcium phosphate.

Yellow : antimony sulphide, Fluorescent glass : uranium

silver borate, or selenium. oxide.

Violet : manganese dioxide. Black glass : ferric oxide and

cupric oxide.

STRONTIUM, Sr = 86-93, AND BARIUM, Ba = 136-28.

Strontium and barium minerals. — The mineral heavy spar, or
bargtes (Greek baros = heavy, from its high density, 4-5), is a very
common vein stone in lead mines, where it is associated with galena,
calcite, fluorite, and quartz, and is called cawlc by the miners. In
1602, Vincentius Casciorolus, a shoemaker of Bologna, found that if

XL THE ALKALINE -EARTH METALS 851

barytes is ignited with charcoal, the residue is phosphorescent after
exposure to light. Barium sulphide is formed by reduction of the
sulphate (barytes) : BaSO4 + 40 = - BaS + 4CO. In 1774, Scheele
examined barytes, and concluded that it was the sulphate of a pecu-
liar earth, called baryta by Lavoisier. Barium also occurs as the
carbonate, BaC03, the mineral witherite, a gangue material in lead
veins.

A peculiar mineral found in the lead mines of Strontian, in Argyle-
shire, was examined by Hope in 1791, and by Klaproth in 1793.
They concluded that it was the carbonate of a new earth, different
from lime and baryta, to which Klaproth gave the name of strontia.
The mineral, called strontianite, is strontium carbonate, SrCO3.
Strontium sulphate, SrS04, occurs as the mineral celestine, so called
from the pale blue colour of some specimens.

Strontium and barium salts. — If the carbonates are dissolved in
hydrochloric acid, the iron in the solutions oxidised with chlorine
water, precipitated by boiling with a little of the strontium or barium
carbonates obtained by adding sodium carbonate to a portion of the
solution, and the filtered liquid evaporated, crystals of strontium
chloride, SrCl2,6H20, or barium chloride, BaCl2,2H2O, are formed.
The former are efflorescent, but the latter are unchanged in the
air. Strontium chloride is soluble in alcohol, whilst barium chloride
is insoluble. This property may be used to separate calcium and
strontium (chlorides soluble in alcohol) from barium (chloride
insoluble) i

By carrying out the above operation with dilute nitric, instead
of with dilute hydrochloric, acid, strontium nitrate, Sr(NO3)2, and
barium nitrate, Ba(N03)2, are formed. These salts are used in
pyrotechny to produce crimson and green fire, respectively, by
mixing them with sulphur and charcoal. They are insoluble in
alcohol (calcium nitrate is soluble);

To prepare soluble salts from the natural mineral sulphates, which
are sparingly (SrSO4), or not at all (BaSO4), soluble in acids, they may
be fused with sodium carbonate, when the carbonates are produced,
and may be separated from the alkali sulphate by boiling the mass
with water and washing : BaS04 + Na2C03 = BaCO3 + Na2SO4.
In another process the minerals are strongly heated with carbon,
when the sulphides are formed : BaSO4 + 40 = BaS + 4CO. The
carbonates or sulphides may then be dissolved in acids, and the salts
crystallised.

Strontium carbonate is decomposed at a higher temperature than
calcium caabonate, whilst barium carbonate is stable at a bright red
heat. If, however, the carbonates are mixed with charcoal, heated
to redness, and exposed to a current of steam, the hydroxides are
formed : BaC08 + C + H2O - Ba(OH). + 2CO. The oxides are
best prepared by heating the nitrates. They are white, amorphous

3 I 2

852 INORGANIC CHEMISTRY CHAP.

substances resembling quicklime, and combine with water to
form hydroxides with evolution of heat. Strontium hydroxide,
Sr(OH)2,8H2O, is crystalline, and dissolves fairly readily in hot water ;
on heating to redness, it loses water and leaves strontia, SrO. Barium
hydroxide also forms a crystalline hydrate, Ba(OH)2,8H2O, which
dissolves readily in hot water. On exposure to air free from carbon
dioxide the crystals effloresce, forming Ba(OH)2,H20. Barium
hydroxide fuses on heating strongly, but does not lose water even
at a very high temperature ; baryta, BaO, is prepared by igniting the
nitrate. A solution of barium hydroxide in water is called baryta-
water, and gives a white precipitate of the carbonate, BaC03, with
carbon dioxide.

Baryta in solution is a strong base, and is often used in volumetric
analysis instead of caustic soda, since any carbonate formed by
exposure to air is precipitated and does not remain in solution to
interfere with the colour-changes of indicators. Barium salts
are poisonous.

Baryta and strontia, as well as lime, form sparingly soluble
compounds, called saccharates, with cane-sugar. These are used
in separating the sugar from the molasses residues of beet-sugar
(which are not palatable) ; since barium compounds are
poisonous, the strontium or calcium compounds are used in practice :
C12H22Ou,2SrO ; CuHiBOn,3C5aO. The precipitates, after wash-
ing, are suspended in water and, decomposed by a current of carbon
dioxide. The carbonate is precipitated, pure sugar remains in
solution, and may be crystallised.

Barium peroxide, Ba02, is obtained by passing oxygen or air over
baryta heated to dull redness : 2BaO + 02 ^ 2Ba02. The dissocia-
tion pressures at different temperatures are :

555° 650° 720° 790° 795°
25 65 210 670 760 mm.

Strontium peroxide, Sr02, is formed from the monoxide and oxygen
at a dull red heat under a pressure of 125 kgm./sq.cm., and is
similar to barium peroxide. (Calcium peroxide has not yet been
obtained directly.) The hydrates of barium and strontium peroxides
BaO2,8H20 and SrO2,8H2O, are obtained as crystalline precipitates,
by adding hydrogen peroxide to cold saturated solutions of barium
and strontium hydroxides. On gently heating, the hydrates lose
water and form Ba02 and Sr02. By precipitating a concentrated
solution of the hydroxide above 40°, anhydrous Sr02 is formed ;
below 40° the compounds Ba02,H202 and Ba02,2H202 are formed
with excess of hydrogen peroxide and baryta-water.

Metallic strontium and barium are obtained by the electrolysis of
the fused chlorides, preferably mixed with potassium chloride, or by
heating the oxides with aluminium powder in vacuo. They are

XL THE ALKALINE -EARTH METALS 853

silver-white. Strontium (sp. gr. 2-5) melts at 900°, barium
(sp. gr. 3-6) at 850°. Barium is also obtained by heating the oxide
with silicon in an evacuated steel tube : 3BaO + Si = BaSiO3 -f
2Ba. The metal distils off.

Barium sulphate, BaSO4, is formed as a fine white precipitate
insoluble in acids (except hot concentrated sulphuric acid, from which
crystals of the acid sulphate, Ba(HS04)2, separate on cooling, or in
hot very concentrated hydrochloric acid), by adding sulphuric
acid or a sulphate to a barium salt. It is used as a pigment (per-
manent white), but has a poor covering power. Lithopone is a
mixture of BaSO4 and zinc sulphide made by precipitation :
BaS -f ZnS04 = BaSO4 -f ZnS. It has a better covering power
even than white lead (p. 928), and does not darken on exposure to
sulphuretted hydrogen.

EXERCISES ON CHAPTER XL

1. In what forms does calcium carbonate occur ? How is quicklime
manufactured from limestone ?

2. Starting from quicklime, how would you prepare calcium peroxide,
chloride, sulphide, and nitrate ?

3. Describe the general properties of the alkaline -earth elements.
Discuss the position of magnesium in the group.

4. How are barium and strontium salts obtained from the minerals ?

5. Describe the preparation of : (a) metallic calcium, (b) barium
fluoride, (c) strontium nitrate from celestine. What are the properties
of these substances ?

6. How are mortar, cement, and glass made ? What is known of the
composition of these materials ?

7. Describe the manufacture of calcium carbide. What important
substances are prepared from it ?

8. What is " superphosphate of lime " ? How is it manufactured ?

CHAPTER XLI

THE METALS OF THE ZINC GROUP

Beryllium, Be = 9-0. — Vauquelin, in 1798, found that the mineral
beryl (Fig. 404) contains a peculiar earth, which he called glucina,
differing from lime and alumina by forming a soluble sulphate which
does not produce alums. The Peruvian emerald (cf. p. 894) is a trans-
parent variety of beryl, coloured green by oxide of chromium. The
formula of beryl is 3BeO,Al2O3,6SiO2. To prepare beryllium salts
from beryl, it is fused with potassium carbonate, the melt evaporated
with sulphuric acid, and digested with water. Silica is filtered off,
and on cooling the evaporated nitrate, nearly all the aluminium sepa-
rates in the form of potash alum. The mother liquor is then poured
into a concentrated solution of ammonium
carbonate. Beryllium hydroxide and carbonate,
Be(OH)2 and BeCO3, are soluble in ammonium
carbonate, whilst ferric hydroxide and alumma
are precipitated. The nitrate on boiling deposits
a basic beryllium carbonate. If this is ignited,
beryllium oxide, or beryllia, BeO, remains as a
white powder soluble in hot concentrated sul-
phuric acid ; the solution on cooling deposits
°f *™M™ Sulphate, BeSO4,4H2O,

FI8.404.-Crystalof .

Beryl. possessing a sweet taste (hence the name

glucinum, often given to the element). The

sulphate does not form mixed crystals with CuSO4, FeSO4, etc., and
thus differs from ZnSO4 and MgSO4. By passing chlorine over a
heated mixture of the oxide and carbon, the chloride, BeCl2, sublimes
in white crystals which fume in moist air. The vapour density of the
chloride at 520° corresponds with the formula BeCl2 (p. 469). Metallic
beryllium is obtained by the electrolysis of a fused mixture of the
chloride with sodium and ammonium chlorides, or of the fluoride with
sodium fluoride in a nickel crucible with a carbon anode. It is a hard,
white metal, sp. gr. 1-842, m.-pt. 1278°, which burns brilliantly in the
air when heated in the form of powder, but does not decompose steam
even at a red heat.

Beryllium hydroxide, Be(OH)2, is soluble in alkalies, but is reprecipi-

854

CH. XLI

THE METALS OF THE ZINC GROUP

855

tated on diluting the solution. It is readily soluble in ammonium
carbonate. By these reactions it is distinguished from alumina, which
it otherwise closely resembles.

MAGNESIUM. Mg = 24*13.

Magnesium. — In 1695, Nehemiah Grew obtained from the water
of a mineral spring at Epsom a peculiar salt which was called
Epsom salts. The salt was afterwards found in other mineral springs,
in the mother liquors from the preparation of common salt from sea-
water, and in saltpetre mother liquors. Epsom salt is magnesium
sulphate, MgS04,7H2O ; magnesium chloride, MgCl2, is contained
in sea- water. By precipitating solutions of these salts with potassium
or sodium carbonate, magnesia alba, which like Epsom salt is used
medicinally, is obtained. Black, in 1755, showed that magnesia
alba is a compound of fixed air, or carbon dioxide, with calcined
magnesia, or magnesium oxide, MgO, left after ignition of magnesia
alba. Metallic magnesium was obtained in an impure state by Davy.

Magnesium is widely distributed, occurring in the forms of
magnesite, MgC03; dolomite, MgCO3,CaCO3; kieserite, MgSO4,H20 ;
kainite, MgSO4,K2S04,MgCl2,6H2O ; and carnallite, KCl,MgCl2,6H20.
It is also contained in spinel, MgO,Al2O3, and is an important con-
stituent of rocks: olivine, Mg2SiO4; talc, Mg3H2 (SiO3)4 ; asbestos,
CaMg3(Si03)4 ; meerscliaum, H2Mg2(Si03)3,H20, augite, olivine, and
serpentine (p. 746) are common rock-forming minerals. All plant-
and animal -tissues contain magnesium ; it appears to be an
essential constituent of chlorophyll, the green
colouring-matter of plants (p. 694).

Magnesium salts considerably in excess of
demand are obtained as by-products at Stass-
furt (p. 790).

Magnesium sulphate, MgS04. — Magnesium
sulphate is prepared from magnesite, MgCO3, or
dolomite, MgC03,CaCO3. Magnesite occurs in
large masses in various localities, e.g., in Greece.
If magnesite or dolomite is boiled with dilute
sulphuric acid, calcium carbonate is converted
into the sparingly soluble gypsum, and mag-
nesium sulphate goes into solution. Iron is
separated by boiling with a little precipitated
magnesium carbonate, and the filtrate on evap-
oration and cooling yield:-} crystals of magnesium sulphate, MgS04,7H20
(Epsom salts] (Fig. 405). These are also formed by dissolving
kieserite in boiling water (it is practically insoluble in cold water),
and crystallising. Magnesium sulphate is used as a purgative, as a
dressing for cotton goods, and in dyeing with aniline colours.

FIG. 405.— Epsom Salt
Crystal.

856 INORGANIC CHEMISTRY CHAP.

Several hydrates of MgSO4 are known, e*g., with 7H2O, 6H2O, and
H2O ; at 200°, the anhydrous sulphate is formed from the hydrates.
The hydrate MgSO4,7H2O is formed from supersaturated solutions on
cooling ; it is isomorphous with ZnSO4,7H2O. Double salts with
alkali-metals are readily formed, e.g., MgSO4,K2SO4,6H2O is schdnite,
a Stassfurt mineral. A solution of the anhydrous sulphate in con-
centrated sulphuric acid deposits crystals of MgSO4,H2O.

The double salts in solution are completely decomposed into the
single salts, as is shown by the magnitude of the depression of
freezing point. They are in this way distinguished from complex
salts, such as K4Fe(CN)6, which retain their constitution in solution,
and ionise accordingly : K4Fe(CN)6 — 4K' + Fe(CN)6"". The
solution then does not exhibit the reactions of the components
of the complex ion (e.g., Fe", and CN'). Isomorphous mixtures,
or mixed crystals, e.g., a mixture of FeS04,7H2O and MgS04,7H20,
differ from double salts by having a variable composition. They
may be represented by such formulae as (Fe,Mg)SO4,7H20 (cf. p. 446).

Magnesium chloride, MgCl2. — Carnallite, KCl,MgCl2,6H2O, occurs
in large quantities in the Stassfurt deposits (p. 790). It fuses at
176°, undergoing decomposition with deposition of practically all
the potassium chloride. Fused magnesium chloride, MgCl2,6H2O,
remains. On cooling this the rest of the potassium deposits as
carnallite, and the fused residue of magnesium chloride solidifies
to a crystalline mass of MgCl2,6H2O. The crystals are very deli-
quescent, and are used in lubricating cotton thread in spinning.
Magnesium chloride forms several hydrates, viz., with 12H20,
8H2O (a and 0), 6H20, 4H20, and 2H20. If the crystalline
hydrates are heated above 186° they undergo hydrolysis : hydro-
chloric acid and steam are evolved, and magnesium oxide is left :
MgCl2 -f- H2O = MgO 4- 2HC1. Anhydrous magnesium chloride,
MgCl2, is prepared by heating the hexahydrate in a vacuum at 175°,
or in a current of hydrogen chloride. Another method is to add
ammonium chloride to the solution, evaporate, and heat in a
covered crucible. The double salt, MgCl2,NH4Cl,6H20, loses water,
and the residual MgCl2,NH4Cl on further ignition evolves a mixture
of hydrogen chloride and ammonia, leaving fused anhydrous
magnesium chloride, MgCl2.

The hydrolysis is prevented by the production of the stable com-
pound MgCl2,NH4Cl, from which the water may be completely removed
at a temperature below that at which decomposition occurs. Mag-
nesium bromide, MgBr2,8H2O, and iodide, MgI2,8H2O, occur in some
mineral springs, and are prepared in the same way as the chloride, by
dissolving magnesium oxide or carbonate in the acids.

If a concentrated solution of magnesium chloride is mixed with

XLI THE METALS OF THE ZINC GROUP 857

magnesium oxide, the paste solidifies to a hard, white mass of the
oxychloride, Mg(OH)Cl. This is used as a dental stopping, and as
a finish for plaster, since it takes a fine polish.

Magnesium. — Metallic magnesium is prepared by the electrolysis
of fused carnallite, which loses water and fuses to a clear liquid at
700°. Calcium fluoride is also added. The cathode is the iron
crucible, the anode is of carbon. The chlorine is led off, and the
metal floats to the surface, being protected by a current of coal gas.
The semi-fused metal is pressed into wire, which is then rolled into
ribbon. It appears to be prepared in England by Vickers by tjie
old expensive process of reducing the fused double chloride of
magnesium and sodium with metallic sodium. Magnesium may
also be prepared by the electrolysis of a solution of magnesium
ammonium sulphate at 70-100°. Metallic magnesium in the form
of ribbon burns when heated in air with an intense white light,
producing the oxide, MgO, and a little nitride, Mg3N2. The residue,
when moistened with water, therefore gives off a little ammonia.
Magnesium powder mixed with powdered potassium chlorate or
barium peroxide burns explosively when lighted, producing a
blinding white flash. The mixture is used in photography, and for
signalling and star-shells. A mixture of magnesium and dry
amorphous silica is also used. The metal is stable in dry air, but
soon becomes covered with oxide in moist air : the alloys with
lead, containing Mg2Pb, rapidly oxidise in air. Magnesium melts
at 651°, and boils at 1100°. It is very light (sp. gr. 1-75). Fine
crystals are formed by subliming the metal in an evacuated tube
at about 550°. The metal dissolves readily in dilute acids, but not
in alkalies. Magnesium powder decomposes water (p. 182). A
colloidal solution in ether can be prepared by Bredig's method.

Magnesium combines directly with nitrogen on heating, forming a
nitride, Mg3N2, a greenish-yellow, amorphous mass, decomposed by
water : Mg3N2 + 3H2O = 3MgO + 2NH3. A sulphide, MgS, two
carbides, MgC2 and MgC3, and two silicides, Mg2Si and MgSi, are formed
by direct combination. The sulphide is at once hydrolysed by water,
although a solution, probably containing the hydrosulphide, Mg(HS)2,
is formed by passing sulphuretted hydrogen through the oxide suspended
in water. It decomposes on warming, evolving pure hydrogen sulphide.
The phosphide, Mg3P2, and arsenide, Mg3As2, are formed by direct
combination and are decomposed by water.

Magnesia, MgO. — By precipitating a solution of magnesium
sulphate or chloride with caustic soda, and drying at 100°, the
sparingly soluble hydroxide, Mg(OH)2, is formed, insoluble in
excess of alkali. This occurs crystalline as the mineral brucite.
On heating, the hydroxide loses water and forms the oxide, MgO,
which occurs in octahedral crystals as periclase. Magnesium oxide

858 INORGANIC CHEMISTRY CHAP.

is usually prepared by heating the basic carbonate (q.v.), or native
magnesite, and is known as calcined magnesia. Two varieties are
formed, from the light and heavy carbonates respectively, the
specific gravities of which are in the ratio 1 : 3-5. The oxide slowly
combines with water, forming the hydroxide, and when moist
turns red litmus paper blue. It fuses at about 2250°, and is reduced
by carbon in the electric furnace, forming magnesium carbide. A
crystalline form is produced on heating the powder strongly hi a
current of hydrogen chloride.

Magnesia, prepared by the calcination of native magnesite, is used
in the manufacture of refractory bricks for electric furnace -linings.
These are basic, and resist the action of basic slags containing lime.

Acidic linings are composed of ganister (largely silica), and neutral
linings of chromite or chrome -ironstone (p. 947). Bricks containing
90 parts of MgO, 5FeO, and 5 of silica, chalk, and clay, sinter above

1400°, but do not fuse below 2000°.

The solubility of magnesium hydroxide (1 part in 55,000 of water)
is reduced by the addition of potash or soda, in accordance with
the equation Mg(OH)2 ^± Mg- -f- 2OH', but is increased by the
addition of ammonia, and especially of ammonium chloride. This
reaction is applied in qualitative analysis, where magnesium is
kept in solution by ammonium chloride whilst the metals of the
groups III, IV, and V are precipitated by NH4OH, NH4HS, and
(NH4)2C03, respectively.

The solubility of magnesium hydroxide in ammonium salts is due
to the feeble ionisation of ammonium hydroxide, NH4OH. If an
ammonium salt is brought in contact with Mg(OH)2, the OH' ions of
the latter are withdrawn from the solution to form practically un-ionised
NH4OH, the ionisation of which is still further reduced by the excess
of NH4' ions of the NH4C1. More Mg(OH)2 therefore dissolves and
dissociates, to provide a further supply of OH' ions, and the process
goes on until the solubility product [Mg-] x [OH']2 is reached, or if
this cannot be attained, until all the Mg(OH)2 is dissolved.

A peroxide, probably MgO2, is obtained in an impure state by
precipitating a solution of the sulphate, mixed with hydrogen per-
oxide, with caustic soda. After drying, it contains about 8 per cent,
of available oxygen, and is used as an antiseptic in tooth-pastes,
etc.

Magnesium carbonates. — The normal carbonate, MgC03, occurs
native as magnesite. The precipitates formed by adding car-
bonates to solutions of magnesium salts are always basic carbonates.
Fr9m solutions at the ordinary temperature, the precipitate is
a light loose powder — magnesia alba levis, of the composition
4MgCO3 -j- Mg(OH)2 -f 5H26. From a boiling solution, a denser

XLI THE METALS OF THE ZINC GROUP 859

crystalline precipitate, MgC03 -f Mg(OH)2 + 4H2O, magnesia alba
ponder osa, is thrown down. Both are used in medicine. If the
basic salts are suspended in water, and a current of carbon dioxide
is passed through, they dissolve, producing a bicarbonate. The solu-
tion containing 2-65 gm. of Mg(HCO3)2 per 100 c.c., is known as
fluid magnesia. If the solution is heated to 50°, crystals of mag-
nesium carbonate, MgC03,3H2O, separate.

Magnesium phosphate, Mg3(P04)2. — This compound occurs in
bones and in plant-tissues, and is precipitated from solutions of
magnesium salts by trisodium phosphate, Na3P04. Ordinary
sodium phosphate, Na2HPO4, precipitates magnesium hydrogen
phosphate, MgHP04, soluble in 322 parts of cold water. On heating
the solution, the normal salt, Mg3(P04)2, is precipitated, and an acid
salt, supposed to be MgH4(P04)2, remains dissolved. If a solution
of a magnesium salt is mixed with solutions of ammonium chloride
and ammonia, and a phosphate added, a crystalline precipitate is
deposited, slowly from dilute solutions, but more rapidly on stirring
or scratching the sides of the beaker with a glass rod. This consists
of magnesium ammonium phosphate, Mg(NH4)P04,6H20. This
substance is present in some urinary calculi ; it is sparingly soluble
in water (1 part in 15,000), and less so in dilute ammonia (1 part in
44,000) ; its formation is a delicate test for a phosphate (p. 632),
or magnesium.* On ignition, it is converted into the pyrophosphate,
Mg2P2O7, in which form magnesium is estimated in gravimetric
analysis : 2Mg(NH4)PO4 = Mg2P2O7 + H2O + 2NH3.

Magnesium is separated from the alkalies by adding baryta-water,
when Mg(OH)2 is precipitated. The excess of baryta is precipitated
from the filtrate by saturation with carbon dioxide, when BaCO3 is
formed, leaving the alkali carbonates in solution. The precipitate of
Mg(OH)2 is washed, dissolved in dilute hydrochloric acid, and pre-
cipitated as MgNH4PO4.

The atomic weight of magnesium was determined by the analysis
of the anhydrous chloride, prepared by heating MgCL,NH4Cl in a
current of HC1 gas. P is 24-13 (H = 1).

ZINC. Zn == 64 »85.

Zinc minerals. — The ancients prepared orichalcum, or brass, an
alloy of zinc and copper, by heating copper with an ore known as
cadmia and charcoal. Cadmia, which was called tutia, or tutty,
by the alchemists, was probably zinc carbonate, ZnC03. Since
the copper was turned a golden-yellow colour by this process, tutia

* Magnesia mixture, for the precipitation of phosphates, is prepared by
dissolving 50 gm. of MgCl2,6H2O and 75 gm. of NH4C1 in 1 litre of 4 per
cent, ammonia solution.

860 INORGANIC CHEMISTRY CHAP.

was looked upon as an approach to the Philosopher's Stone. Thus,
Geber says : copper " agrees very well with Tutia, which citrinizeth
it with good yellowness ; and hence you may reap profit. Therefore
take it, before all other Imperfect Bodies, in the Lesser and Middle
Work, but not in the Greater."

The metal itself is first mentioned by Paracelsus, who refers to it
as zinken — a " semi-metal " or "a spurious son of copper " (pro-
bably on account of its brittle character). The name spelter for the
metal is used by Boyle, but was also applied to bismuth, with which
zinc was confused. Libavius describes the metal, which he says
was brought from the East Indies. The real nature of brass was not
clear until Kunckel observed that : " calamine allows its mer-
curial [i.e., metallic] part to pass into the copper and form brass."
Zinc was identified as the metal from blende (ZnS) by Homberg in
1695 ; the extraction of the metal from calamine was effected by
Lawson early in the following century.

Zinc occurs chiefly as blende, ZnS, usually coloured yellow or
brown by iron (" black-jack " of the miners), and possessing a
characteristic resinous lustre. It is found in England, in many-
parts of Europe and America, and in New South Wales. The
carbonate occurs as calamine, or smithsonite, ZnC03, in Belgium,
Germany, and America. Electric calamine is a silicate, Zn2Si04,H2O.
The oxide, zincite, ZnO, is a rare ore, but the ferrite, Zn(Fe02)2, or
ZnO,Fe203, forms the very important franklinite, or red zinc ore,
deposit "of Franklin Furnace, New Jersey. The New South Wales
ore contains galena, and is first " concentrated " by the flotation
process (p. lO). Certain varieties of pyrites, e.g., Westphalian,
contain zinc sulphide. Traces of zinc occur, as an organic com-
pound, in animal cells, and especially in snake venom (0-11-0-50
per cent.).

Metallurgy of zinc. — The extraction of zinc from its ores was in
operation on an extensive scale at Bristol in 1743, the roasted ore
(ZnO) being distilled with carbon at a high temperature in a crucible
the bottom of which was perforated and fitted with a piece of iron
pipe passing above the surface of the mixture inside. Zinc is a
volatile metal (b.-pt. 918°), and distilled off, the vapour condensing
in the lower part of the tube to liquid metal, which ran into water.
This process is no longer used. In 1807 zinc smelting was begun at
Liege, in Belgium, and later on spread to Silesia. These two pro-
cesses are still in use, and are called the Belgian process, and the
Silesian process, respectively. In America, franklinite ore is used ;
in Europe, blende.

The ore is first roasted, and the sulphur dioxide produced from
blende may be utilised in the manufacture of sulphuric acid. Exter-
nal heating has to be applied, the ore being raked in a series of
muffles, through which air circulates. Care must be taken th*

XLI

THE METALS OF THE ZINC GROUP

861

only the oxide is produced : 2ZnS + 302 = 2ZnO + 2S02, since
the sulphate, ZnS04, if formed, is very stable, and would, in the
subsequent reduction, again give sulphide, leading to considerable
loss. The roasted ore is next mixed with half its weight of powdered
coal and charged into fireclay retorts, which are strongly heated.
Zinc distils off : ZnO+C = Zn+CO. The reduction begins at 800°,
and increases rapidly with rise of temperature. The Belgian retorts
consist of fireclay tubes, closed at one end and set in a furnace,
sloping towards the open end (Fig. 406). An iron tube is luted into
the open end with clay and serves to condense the zinc. The
Silesian retorts are fireclay muffles (Fig. 407), to which a fireclay
elbow and an iron con-
denser tube are luted ;
they are heated in a fur-
nace. The newer fur-
naces employ three rows
of muffles, one above the
other, the lower row being
supported along their
length on the hearth,
and the two upper rows
only at the ends. They
are fired with gas. In
all cases 10-25 per cent,
of the zinc is lost. The
production in 1909 was
800,000 tons.

The ore is also smelted
to a limited extent in the
electric furnace, either of
the arc or resistance type,
but a considerable pro-
portion of the metal is
then obtained in the form
of a powder mixed with oxide, known as zinc dust, which is also
produced, to a less extent, in the coal-fired furnaces.

Commercial zinc, called spelter, contains about 97-98 per cent,
of zinc, 1-3 per cent, of lead, and some arsenic. It is refined by
electrolysis in an acid solution of zinc sulphate with a high current
density (Pring and Tainton). This metal contains 99-9 per cent, of
zinc. Zinc is also leached from burnt pyrites containing it, with
ferric sulphate solution, and deposited electrolytically. The impure
metal dissolves readily in dilute acid, whereas some varieties of
the pure metal dissolve slowly unless a few drops of copper sulphate
or platinic chloride solution are added. Metallic copper, or plati-
num is precipitated on the zinc, and forms a galvanic couple, from

FIG. 406.— Belgian Zinc Furnace.

862 INORGANIC CHEMISTRY CHAP.

the insoluble part of which hydrogen is readily evolved. For use
in the laboratory the metal is usually granulated by melting in a
clay crucible and pouring into a bucket of water. Zinc foil, or
sheet, is prepared by heating the metal to 100-150°, when it becomes
soft, and rolling it.

Metallic zinc. — Zinc has a bluish-white colour, melts at 419°, and
boils at 918°. Its vapour density corresponds with the formula
Zn. The metal is hard and moderately brittle ; it softens at 100-
150°, but becomes very brittle at 205°, and can then be powdered
in a mortar. It readily burns in air when the turnings are heated
in a flame, or the metal is heated strongly in a crucible, producing
a white cloud of oxide, which settles out in the form of woolly
flocks. These were called " Philosophers' wool " by the alchemists,

FIG. 407. — Silesian Zinc Furnace.

or, in Latin, nix' alba (white snow). This name was rendered as
" Weisses Nichts " into German, and thence, by Teutonic erudition,
nihilum album. The metal oxidises in moist air, forming the basic
carbonate, and is attacked and dissolved by soft water, especially
that containing peat acids. The zinco-solvency of water is reduced
by allowing it to stand over limestone.

Zinc and copper are the constituents of the valuable alloy brass.
Zinc is miscible in a state of fusion with tin, copper, and antimony,
but only partly dissolves in lead and bismuth (cf. p. 820).

Zinc is more resistant to moist air than, and is used as a protection
for, iron. The iron sheets or wire are cleaned by a sand-blast,
and dipped into molten zinc, when an adherent coating of the latter
is formed. This process is known as galvanising, and the product

XLI THE METALS OF THE ZINC GROUP 863

as galvanised iron. Iron articles may also be coated with zinc by
heating them in zinc dust ; this is known as sherardising. The
zinc dissolves before iron in presence of oxygen and moisture, since
it is more electropositive than the other metal. The metal is also
used for the positive electrodes of voltaic cells (p. 881).

Zinc dissolves in dilute acids, evolving hydrogen (except with
nitric acid), and producing zinc salts (p. 185), containing the zinc
ion, Zn". It also dissolves readily in hot solutions of caustic potash
and soda, evolving hydrogen, and forming solutions of zincates :
Zn + 2KOH = K2Zn02 + H2.

Zinc oxide, ZnO. — Zinc oxide, ZnO, is produced by the combustion
of the metal ; when so prepared it is called zinc white and is used
as a pigment. It is prepared for pharmaceutical purposes by pre-
cipitating a solution of zinc sulphate with sodium carbonate, and
igniting the basic carbonate. It is a white powder which becomes
sulphur-yellow on heating, the colour disappearing on cooling.
Zinc oxide sublimes readily at 1400°. On exposure to air it takes
up a little water. Zinc oxide dissolves readily in acids, producing
zinc salts, and in alkalies, forming zincates, e.g., KHZn02, and
NaHZnO2,3H20, which can be obtained in the solid state.

Zinc hydroxide, Zn(OH)2, is formed as a white, flocculent pre-
cipitate on adding caustic potash or soda to a solution of a zinc
salt. It can be dried at 85°, but loses water at higher temperatures.
Zinc hydroxide dissolves in 190,000 parts of water. The precipitate
is readily soluble in excess of the alkali, producing a solution con-
taining colloidal zinc hydroxide and a little zincate. Zinc hydroxide
is therefore feebly acidic as well as basic ; it is an amphoteric sub-
stance. Ammonia also dissolves it, forming a complex hydroxide :
Zn(NH3)4(OH)2 — Zn(NH8)4'* + 20H'.

By the action of 30 per cent, hydrogen peroxide on zinc oxide
at —10°, a white, or yellow, powder is obtained, which is believed
to be a hydrated peroxide, ZnO2,Aq. By the action of 30 per cent.
II202 on a solution of zinc oxide in caustic soda (sodium zincate), a
white precipitate of the formula Zn02,H2O, or ZnO,H202, is formed.
Precipitates obtained by adding zinc sulphate to solutions of Na202
are probably mixtures of zinc hydroxide and peroxide.

Zinc oxide is used as an absorbent in surgical dressing, as a
" filling " for rubber, and in the preparation of Rinman's green.
The latter is obtained by heating zinc oxide with a solution of
cobalt nitrate, and is either cobalt zincate, CoZnO2, or a solid solu-
tion of cobalt oxide in zinc oxide. The formation of this green
substance is the basis of the blowpipe test for zinc.

Zinc chloride, ZnCl2.— Anhydrous zinc chloride (b.-pt. 730°) is
formed by passing hydrogen chloride over heated zinc, or by dis-
tilling the metal with mercuric chloride : HgCl2 -f- Zn = Hg + ZnCl2.
It is formed in solution by dissolving zinc or its oxide in concen-

864 INORGANIC CHEMISTRY CHAP.

trated hydrochloric acid (Glauber, 1648 — oil of calamine). On
evaporation, a syrupy liquid is obtained ; if a little concentrated
hydrochloric acid is added to this, small, deliquescent crystals,
ZnCl2,H2O, separate. If, however, the aqueous solution is evapo-
rated to dryness, the oxychlorides Zn(OH)Cl and Zn2OCl2, are formed.
An oxychloride is also produced by mixing the syrupy solution of
the chloride with zinc oxide and finely powdered glass ; the whole
sets rapidly to a very hard mass, used as a dental stopping. The
concentrated solution of zinc chloride is used for impregnating
timber to prevent its destruction by micro-organisms (" dry rot "),
and as a caustic (it dissolves proteins). In timber-preserving, zinc
chloride is being replaced by fluorides. By evaporation in a current
of hydrochloric acid gas the fused salt is obtained, which may be
cast into sticks.

A solution of zinc chloride prepared by adding zinc to commercial
hydrochloric acid (spirits of salt) is used under the name of " killed
spirits " as a flux in soldering. On heating, it liberates hydrochloric
acid, which dissolves metallic oxides and keeps the metal surface clean.
Hot zinc chloride solution dissolves cellulose, forming a colloidal
solution. If this is squirted into alcohol, a thread of amorphous
cellulose is formed, which is carbonised by heating, and forms the
carbon filament of electric lamps. Zinc chloride is used, like mag-
nesium chloride, for " filling " (i.e., weighting and adulterating)
cotton goods. The double salts, ZnCl2,2NH4Cl and ZnCl2,3NH4Cl,
are formed as crystals in Leclanch6 batteries, and zinc chloride absorbs
ammonia gas. The double salts are hydrolysed by water, with depo-
sition of white oxychlorides, which dissolve in dilute hydrochloric acid.

The bromide and iodide, ZnX2, are formed from the elements in
presence of water.

Zinc sulphate, ZnS04. — Zinc sulphate, ZnS04,7H20, isomorphous
with Epsom salts, is known as white vitriol. It was described
by Basil Valentine (p. 29), and was produced by lixiviating
roasted blende. Its composition was correctly given by Neumann
(1735). The substance is the commonest salt of zinc, and is pre-
pared by dissolving the metal, oxide, or carbonate in dilute sul-
phuric acid and evaporating (p. 185). It forms several hydrates ;
on heating ZnS04,7H2O to 100°, ZnS04,H20 is left, which loses
water only at a dull red heat. When strongly heated, sulphur
trioxide is evolved and zinc oxide remains. Double salts, e.g.,
K2S04,ZnS04,6H20, are easily prepared. A solution of white
vitriol (J per cent. ZnS04) is used as an eye lotion, and the sulphate
is used in the manufacture of lithopone (p. 853).

Zinc sulphide, ZnS. — Zinc sulphide, ZnS, occurs aa blende, which
is phosphorescent on heating, and exhibits luminous effects on
exposure to a-rays and J£-rays. An artificial phosphorescent

XLI THE METALS OF THE ZINC GROUP 865

sulphide (Sidot's blende) is formed on heating the precipitated
sulphide to whiteness in a covered crucible ; it is used in making
phosphorescent screens for JC-ray and radioactivity work. Per-
fectly pure zinc sulphide is not phosphorescent ; the property .is
conferred by traces of sulphides of heavy metals (bismuth, copper,
manganese). Zinc sulphide is obtained as a white precipitate on
adding ammonium sulphide'to a solution of a zinc salt ; it dissolves
in all dilute mineral acids, but not in acetic acid (cf. MnS). If
sulphuretted hydrogen is passed through a solution of zinc sulphate,
zinc sulphide is at first precipitated, but owing to the acid formed
the precipitation soon ceases : ZnS -j- 2H' — Zn'* -f H2S. If
sodium acetate is added to the solution, the concentration of hydro-
gen ions is kept low by the formation of the very weak acetic acid :
C2H3O2' -f- H* ^z± C2H402. If nickel and cobalt are present, they
are precipitated only after all the zinc has been thrown down.

Zinc carbonate, ZnC03. — Sodium carbonate precipitates a basic
carbonate from a solution of a zinc salt, the composition depending
on the concentrations and temperature. A solution of a bicarbonate,
e.g., NaHC03, however, gives a white precipitate of zinc carbonate,
ZnC03. This is soluble in a concentrated solution of potassium
carbonate, but is precipitated on dilution. When boiled with
sodium carbonate solution, the carbonate, or basic carbonates, form
zinc oxide. Zinc, or zinc oxide, dissolves in water containing CO2.

The cyanide, Zn(CN)2, is formed as a white precipitate by precipitating
zinc acetate with aqueous hydrocyanic acid. It is soluble in potassium
cyanide, forming a complex salt, K2Zn(CN)4 ^ 2K"+ Zn(CN)4".

Zinc nitrate, Zn(NO3)2,6H2O, is a deliquescent salt, soluble in alcohol.
Zincamide, Zn(NH2)2, is formed by the action of ammonia on zinc
ethyl (q.v.) : Zn(C2H5)2 + 2NH3 = Zn(NH2)2 + 2C2H6. On heating to
dull redness it forms the nitride, Zn3N2, a green powder vigorously
decomposed by water : Zn3N2 -f 3H2O = 3ZnO + 2NH3. The phos-
phide, Zn3P2, is a grey mass formed by direct combination of the
elements on heating.

Zinc ethyl, Zn(C2H5)2, is formed as a volatile, spontaneously in-
flammable liquid, by heating zinc with ethyl iodide and then distilling.
Zinc ethyl iodide, Zii(C2H5)I, is first produced as a crystalline com-
pound, which decomposes on heating : 2Zn(C2H5)I = Zn(C2H5)2 + ZnI2.

Complex ammonia compounds are formed with zinc salts, similar
to those of copper, e.g., Zn(NH3)4Cl2,H2O, Zn(NH3)4SO4,H2O,
Zn(NH3)5SO4, etc. (cf. p. 818).

The atomic weight, 64-85 (H = 1), was found by the analysis of the
pure bromide.

Estimation of zinc. — Zinc is estimated by precipitation as basic
carbonate, ignition, and weighing as ZnO, or by electrolysis of an

3 K

866 INORGANIC CHEMISTRY CHAP.'

alkaline solution. In the volumetric method, it may be titrated
with standard potassium ferrocyanide, uranium nitrate (p. 958)
being used as outside indicator : Zn2Fe(CN)6 is precipitated, and excess
of ferrocyanide then gives a brown colour with the uranium salt.

CADMIUM. Cd = 111-51.

Cadmium. — Most zinc ores contain small amounts of another
metal, cadmium, which also occurs as sulphide in the rare mineral
greenockite, CdS. Blende may contain 2-3 per cent, of cadmium,
and calamine up to 3 per cent.

A certain specimen of zinc oxide, which had a yellow colour, although
free from iron, was found by Stromeyer in 1817 to contain the oxide
of a new metal, to which he gave the name cadmium, from cadmia,
the old name for zinc. A similar specimen of zinc oxide used for
pharmaceutical purposes had been confiscated because its solution
gave a yellow precipitate, supposed to be arsenic sulphide, with H2S.
Hermann showed that this was cadmium sulphide.

Cadmium is more volatile than zinc ; the boiling-points of the
metals in the zinc group decrease with rising atomic weight. The
first portions of dust collecting in the receivers of zinc furnaces
in which ores containing cadmium are reduced therefore contain
most of the cadmium, in the form of brown oxide, CdO, mixed
with zinc oxide. The dust is heated strongly with coal in retorts
having long sheet iron cones as adapters. The distillate may
contain 20 per cent, of cadmium, whilst the original oxides contain
only 1-6 per cent. Finally, the product is distilled with charcoal
in small iron or clay retorts.

Metallic cadmium is used as an amalgam as the cathode in the
Weston standard cell. The amalgam is also applied ia dental
stoppings. Cadmium forms very fusible alloys with other^hetals ;
e.g., Wood's fusible metal, m.-pt. 61°, consists of 4 parts of oismuth,
2 each of tin and lead, and 1 of cadmium. Cadmium is a soft,
bluish-white metal, sp. gr. 8-60, melting at 321°, and boiling at 778°.
The vapour density corresponds with the formula Cd. The metal
becomes brittle at 80° ; it is said to exist in two allotropic forms,
with a transition point at 64-9°.

Cadmium compounds. — Cadmium dissolves slowly in dilute
acids, with evolution of hydrogen and formation of cadmium salts,
all of which, except the brown oxide, CdO, and the bright-yellow
sulphide, CdS, are colourless. The hydroxide, Cd(OH)2, is precipi-
tated by caustic soda or potash from the solutions ; it is insoluble
in excess, but dissolves in ammonia, forming a complex hydroxide,
Cd(NH3)4(OH)2. Cadmium is characterised by the readiness with
which it forms complex salts, but this is even more marked in the
case of mercury. Cadmium hydroxide attracts carbon dioxide

XLI THE METALS OF THE ZINC GROUP 867

from the air ; the normal carbonate, CdCO3, is precipitated from
the salts by alkali carbonates (cf. Zn). On heating the hydroxide
or carbonate, or by burning the metal in air, the brown oxide, CdO,
is formed.

Of the soluble salts of cadmium, the sulphate, 3CdS04,8H2O, the
solubility of which is nearly independent of temperature, and the
chloride, 2CdCl2,5H2O, which is efflorescent and is not hydrolysed
by water (cf. ZnCl2), are most important. The peculiar formula?
of the crystalline salts are noteworthy. The halogen salts are all
soluble in water, but they are only very feebly ionised in solution,
forming complex ions in which the metal exists in the negative ion :
2CdI2 ^± Cd-CdI4 ^± Cd'* -f CdI4". Insoluble cadmium salts, e.g.,
CdS, therefore, readily dissolve in a solution of potassium iodide,
since practically all the cadmium ions are removed as complex ions
or un-ionised salts and the solubility product of the former salt is not
exceeded : Cd(OH)2 + 21' - CdI2 + 20H'. If a concentrated
solution of potassium iodide is added to an ammoniacal solution of
a cadmium salt, a white precipitate of Cd(NH3)2I2 is formed. Copper
gives no precipitate. Complex cyanides are easily formed, e.g.,
K2Cd(CN)4. Cadmium iodide is soluble in alcohol, and is used in
photography.

Cadmium sulphide, CdS, is obtained as a bright yellow precipitate,
used by artists under the name of cadmium, by passing sulphuretted
hydrogen through a solution of a cadmium salt which is not too
strongly acid. If the acid concentration exceeds 0-3 normal, the
sulphide is not precipitated : H2S -f- CdSO4 ^± CdS -f- H2S04.

Cadmium is separated from copper by boiling the precipitated
sulphides with dilute sulphuric acid (1 : 5), when CdS dissolves ;
or by adding ammonia to the solution in excess, then potassium
cyanide/ till colourless, and passing H2S ; CdS is precipitated.

Cad*mium dissolves in a hot solution of cadmium chloride and on
pouring into water a white precipitate of cadmous hydroxide, CdOH,
is formed. (The solution probably contains CdCl.) On gently heating
this, yellow cadmous oxide, Cd2O, is obtained. Two other snboxides,
Cd4O and Cd3O2, are said to be formed on heating the oxalate.

Cadmium, the salts of which differ in many ways from those of mag-
nesium and zinc, forms a bridge between these metals (which form only
one series of compounds), and mercury, which forms two series of

I II II

stable compounds, HgX, or Hg2X2, and HgX2.

MERCURY. Hg = 199-0.

Mercury.- -Metallic mercury, which is peculiar in being liquid at
the ordinary temperature, is first mentioned by Aristotle (B.C. 350) ;
Theophrastus (B.C. 300) refers to it as quicksilver : liquid silver

3 K &

868 INORGANIC CHEMISTRY CHAP.

(chutos argyros) ; Dioscorides (1 A.D.) calls it hydrar gyros. Pliny
speaks of native mercury as argentum vivum, and the metal
obtained by heating cinnabar, HgS, its important ore, with charcoal,
as hydrargyrum (liquid silver). The metal was used in the extraction
of gold.

The alchemists regarded mercury as the type of metallic pro-
perties ; all metals, says Geber, are " composed of Argentvive and
Sulphur, pure or impure " (p. 764). " By convenient Preparation
'tis possible to take away such Impurity . . . and supply the
Deficiency in Perfect Bodies." Compounds of mercury, especially
the violent poison corrosive sublimate, HgCl2, first mentioned by
Geber, were used by Paracelsus (1493-1541) and the latrochemists.
Priestley employed a mercury trough in collecting gases which are
soluble in water, and the metal was used by Lavoisier in his famous
experiment on the analysis of air (p. 47). Mercury is used in the
manufacture of barometers and thermometers, and its compounds
corrosive sublimate, calomel (HgCl), and the fulminate, are used in,
the arts and in medicine. The truly metallic character of mercury
does not seem to have been definitely admitted until the metal was
frozen to a malleable solid (m.-pt. -38-8°) by Braune in 1759. It
is readily frozen by a mixture of solid carbon dioxide and ether.

Metallurgy of mercury. — Small quantities of mercury occur
native, or as amalgams and halogen compounds, but the important
ore is cinnabar, mercuric sulphide, HgS, a red mineral found in
Carniola, Hungary, Peru, California, Mexico, Bavaria, China, and
Japan. In the extraction of the metal the cinnabar is roasted
in a current of air : HgS -j- O2 = Hg -f- S02. The metal is not
easily oxidised ; it undergoes only slow oxidation in air at 300°.

In the older process of extraction, now used only at Almaden,
the ore is roasted in a shaft, B (Fig. 408). The ore rests on a per-
forated arch, k, heated below by a fire, A. Air enters through D,
and the vapours pass through six series of openings, /, into series
of stoneware aludels (cf. p. 404), arranged first in a descending and
then in an ascending position on brick arches. The condensed
mercury flows from these into a channel, b, and then into cisterns.
A little mercury vapour passing on is condensed in water, i, in the
chamber, C. The metal is exported in iron bottles with screw
stoppers. The modern furnaces differ according as lump or pow-
dered ore is treated. Lump ore is roasted in admixture with
charcoal in shaft furnaces, the mixture being fed continuously to
the top of the shaft, as in limekilns, and the vapour of the
metal condensed in Y-shaped earthenware pipes, cooled in water.
Powdered ore is treated in Granzita furnaces, consisting of
shafts containing inclined shelves, sloping at an angle of 45° in
alternately opposite directions, over which the ore falls. Flames
and air pass upwards in the opposite direction to the ore and heat

XLI THE METALS OF THE ZINC GROUP 869

the latter. The vapours pass to brick chambers, having cast-iron
water-jackets for cooling, and then to glass and wooden towers.
In these furnaces one ton of ore is worked in forty minutes. The

FIG. 408. — Extraction of Mercury at Almaden.

annual production of mercury is about 3,500 tons, one-third coming
from Spain.

Properties of mercury. — Commercial mercury usually contains
lead and copper. It then leaves a " tail " when allowed to run over
an inclined glass surface, and forms a black scum
of oxides when shaken with air in a stoppered
bottle. The metal is purified by shaking with 5 per
cent, nitric acid containing a little mercurous
nitrate, or running it several times in a thin stream
through this solution in the apparatus shown in
Fig. 409. The metal is then distilled in a
quartz flask under reduced pressure, a slow
stream of air being allowed to bubble through the
metal.

Mercury is a liquid metal with a silver-white
colour. Its density at 0° is 13-5955, and at —185°,
14-383 ; it boils at 357-25°, and the vapour density
corresponds with the formula Hg. The mon-
atomicity of the vapour is proved by the ratio of
specific heats, cp/cv = 1 -667, found by Kundt and
Warburg from the measurement of the velocity of
sound in the vapour at 360° (p. 599). Mercury is
transparent in very thin films, and then transmits
blue light. A colloidal form (hygrol) is obtained by FIG 40J£LPuriflca.
the reduction of mercurous nitrate with stannous turn of' Mercury.

870 INORGANIC CHEMISTRY CHAP.

nitrate in presence of ammonium nitrate : the black precipitate
dissolves in water to a brown solution.

When shaken with different liquids, or triturated with fats or
powders such as sugar, the metal is converted into a grey powder,
consisting of globules which may be as small as 0-002 mm. Grey
mercury ointment is made in this way. The metal is not attacked
by dilute hydrochloric or sulphuric acid, or alkalies, but dissolves
in dilute nitric acid or hot concentrated sulphuric acid.

Mercury dissolves many metals, forming amalgams, which, when
more than a certain amount of metal is present, are solid. Many
of these contain definite compounds, e.g., NaHg2, KHg2. Copper,
silver, lead, gold, etc., are rapidly dissolved by mercury. Iron is
not amalgamated under ordinary conditions, but an amalgam is
formed by triturating iron powder with mercuric chloride and water.
Mercury readily penetrates sheet copper, rendering it brittle. Copper
amalgam becomes plastic when warmed to 100°, and rubbed in a
mortar. After ten to twelve hours it again becomes hard. It is
used for stopping teeth.

EXPT. 325. — Pour a little mercury into a solution of silver nitrate.
A tree like growth of silver amalgam is produced (arbor diance).

Compounds of mercuryl — Mercury forms two series of compounds,
the mercurous compounds, HgX or Hg2X2, and the mercuric com-
pounds, HgXa. The former are obtained with an excess of metal.
Thus, if excess of mercury is triturated with iodine, green mercurous
iodide, Hgl, is obtained ; with excess of iodine, red mercuric iodide,
HgI2, is formed (p. 116). If mercury in excess is treated with cold
dilute nitric acid, mercurous nitrate, HgNO3,H2O, crystallises out,
whilst if mercury is boiled with fairly concentrated nitric acid,
mercuric nitrate, Hg(NO3)2, is formed, which crystallises as
2Hg(NO3)2,H2O, on cooling.

The constitution of the mercurous salts has been the object of several
experiments. H. B. Baker found that the vapour density of carefully
dried mercurous chloride corresponded with the doubled formula Hg2Cl2,
which was also found by Beckmann from the freezing-point of a solution
of mercurous chloride in mercuric chloride. Ogg, from physico-chemical
considerations, also concluded that the mercurous ion has the formula
Hg2". The element, therefore, appears to be always bivalent, the mercuric
compounds being HgX2, whilst the mercurous compounds contain the
group -Hg-Hg-, in which the metal also has a valency of two, and
are therefore analogous to the cuprous compounds, containing -Cu-Cu-
(p. 254). The vapour density of ordinary undried mercurous chloride
corresponds with the formula HgCl, but Harris and Victor Meyer
(1894) showed that the vapour was dissociated into a mixture of Hg
and HgCl2. • If the vapour is contained in a porous earthenware tube,

XLI THE METALS OF THE ZINC GROUP 871

mercury diffuses out, and condenses in globules, whilst the residue
in the tube contains an excess of HgCl2. If a stick of potash is intro-
duced into the vapour, a red coating of mercuric oxide, HgO, is formed
on it, not black mercurous oxide, Hg2O.

The tendency to form complex compounds, which is absent in the
case of magnesium and zinc, but noticeable with cadmium, is very
pronounced in the case of mercury. Numerous stable complex
salts, containing oxygen, sulphur, and nitrogen, are known.

Mercurous compounds. — Mercurous nitrate, HgN03,H2O, is formed
by the action of dilute nitric acid on the metal in the cold, and
readily crystallises from the solution on standing. If water is
added to the crystals, a white precipitate of a basic nitrate is pro-
duced, which redissolves in dilute nitric acid. A little mercury is
usually kept in the solution to prevent oxidation to the mercuric
compound.

Chlorides or hydrochloric acid precipitate white mercurous chloride,
HgCl, from the solution of mercurous nitrate. To obtain a pure
product, excess of chloride is used, and the solution heated. This
salt, called calomel, is of importance in medicine as a purgative.
It is usually prepared by subliming a mixture of mercuric chloride
and metallic mercury, made by triturating the substances in a
mortar. This is heated in an iron pot, and the crust of calomel
formed on the lid is ground to powder and boiled with water to
remove the very poisonous mercuric chloride, some of which always
sublimes unchanged.

Calomel is sparingly soluble in water (04 mgm. per litre at 20° ;
mercurous fluoride, HgF, is soluble in water). It dissolves to some
extent in solutions of chlorides, or concentrated hydrochloric acid,
but is decomposed, with deposition of mercury ; complexes are
formed in solution :

2HgCl + HC1 ^± HHgCl3 + Hg ; HHgCl3 + HC1 ^± H2HgCl4,

giving the ions HgCl3', and HgCl4".

Mercurous bromide, HgBr, is similar to calomel. The iodide, Hgl,
is formed as a green powder by triturating mercury and iodine
(p. 116). On heating, it becomes yellow.

Mercurous sulphate, Hg2S04, is formed by warming an excess of
mercury with concentrated sulphuric acid (or oleum), and deposits
as a coarsely crystalline powder on cooling. It is also formed as a
white precipitate by adding sulphuric acid to a solution of mer-
curous nitrate. When excess of acid is removed by washing,
hydrolysis of the salt commences, and with water at 25° a basic
salt, Hg2S04,Hg20,H20, is formed. Mercurous sulphate is used
as a depolariser in the standard Weston cell, which gives a constant
E.M.F. of 1-0186 volts, nearly independent of temperature, when
made up with pure materials : Cd -f Hg2S04 = CdS04 + 2Hg.

872 INORGANIC CHEMISTRY CHAP.

Mercurous oxide, Hg2O, is formed as a black powder by treating
calomel with caustic soda solution. It decomposes at 100°, or
on exposure to light, into yellow mercuric oxide and metallic
mercury : Hg2O = HgO + Hg.

Mercurous carbonate^ Hg2C03, is precipitated as a yellow powder on
adding excess of potassium bicarbonate to mercurous nitrate solu-
tion, and allowing to stand for a few days to decompose any basic
nitrate. It decomposes at 100° : Hg2C03 = HgO -f Hg + C02,
or on exposure to light.

Mercuric compounds. — The mercuric compounds, HgX2, are
formed by the oxidation of mercurous compounds. Thus, calomel
dissolves in aqua regia, forming mercuric chloride, HgCl2. The
mercuric compounds, conversely, may be reduced to mercurous
compounds, or to metallic mercury. Thus, calomel is precipitated
if sulphur dioxide is passed through a solution of mercuric chloride :
2HgCl2 -f 2H20 + S02 = 2HgCl + 2HC1 + H2S04. By the action
of a solution of stannous chloride, white calomel, or grey finely-divided
mercury, may be precipitated, according to the proportions added :
2HgCl2 + SnCl2 = 2HgCl + SnCl4 ; and with excess of SnCl2 :
2HgCl -f SnCl2 = 2Hg -f- SnCl4. All compounds of mercury are
reduced to the metal if boiled with hydrochloric acid and copper
foil ; the latter becomes white owing to amalgamation, and on
heating the foil in a glass tube a sublimate of minute globules of
mercury is formed. A similar sublimate is obtained directly if a
mercury salt is heated with powdered charcoal and sodium carbonate.

Mercuric nitrate, Hg(N03)2. — This salt is obtained in large, very
deliquescent, colourless crystals, 2Hg(NO3)2,H2O, by boiling mercury
with excess of concentrated nitric acid, cooling, and evaporating
over concentrated sulphuric acid in a desiccator. The mother
liquor on evaporation deposits a basic salt, 2Hg(OH)N03,H2O.
Mercuric nitrate is decomposed by water ; at 25° the basic salt,
Hg(NO3)2,2HgO,H20, is formed as a white powder, decomposed into
oxide by further action of water. Mercuric nitrate is precipitated
by concentrated nitric acid from its aqueous solution.

Mercuric sulphate HgSO4, is obtained by boiling mercury with
one and a half times its weight of concentrated sulphuric acid, and
evaporating to dryness. The white residue may be crystallised
from sulphuric acid. With a small quantity of water, the residue
forms colourless crystals of HgS04,H2O, but it readily hydrolyses,
producing at 25° a basic salt, which is a yellow, crystalline powder,
3HgO,SO3,4H2O, sparingly soluble in water, and called turpeth
mineral. This was described by Basil Valentine.

Mercuric oxide, HgO. — By adding an alkali to a solution of the
nitrate, mercuric oxide, HgO, is precipitated ; from cold solutions
this separates as a yellow, from hot solutions as an orange, powder.
According to Ostwald, the difference in colour is due merely to

XLI THE METALS OF THE ZINC GROUP 873

differences in the fineness of the powder, but Schoch states that the
two varieties have different crystalline forms, and different disso-
ciation pressure at 300°. By heating the nitrate, alone or inti-
mately mixed with mercury, to a moderate temperature, the
crystalline red oxide is formed. A dense red crystalline oxide is
also formed slowly on heating mercury in an open' flask with a long
neck at about 300°. This form, described by the Latin Geber, was
called by the alchemists mercurius prcecipitatus per se, or " red
precipitate." It decomposes on heating (p. 24) ; if the mercury
is kept from condensing, an equilibrium is set up : 2HgO ^
2Hg + 02.

Mercury peroxide, HgO2, is obtained as an amorphous, brick-red
powder when hydrogen peroxide and then alcoholic potash are added
to a solution of mercuric chloride in alcohol. It is fairly stable, but is
decomposed by water. The peroxide is also formed by the action
of H2O2 on HgO, but decomposes with evolution of oxygen, leaving
finely-divided mercury.

Mercuric chloride, HgCl2. — Mercury is rapidly attacked by
chlorine, a white crust of mercuric chloride, HgCl2, forming on the
metal. The action is more rapid if the mercury is heated. Mer-
curic chloride is also called corrosive sublimate on account of its
very poisonous properties, and its volatility (m.-pt. 286° ; b.-pt.
303° ; sp. gr. 541). The fatal dose is 0-2-04 gm. ; the antidote
is the immediate administration of raw whites of eggs, followed
by an emetic. The albumin is coagulated. Corrosive sublimate
is used in preserving skins, as a bactericide, and medicinally : a
O'l per cent, solution is used for sterilising the hands and instru-
ments in surgery.

The preparation of corrosive sublimate is described by the Latin
Geber, who obtained it by subliming a mixture of finely-divided
mercury, calcined green vitriol, common salt, and nitre :

Hg + 2NaCl + 2KNO3 + Fe2S2O9 =

HgCl2 + Na2SO4 + K2SO4 + Fe203 + 2N02.
The use of mercury compounds in medicine was introduced by
Paracelsus, and by the end of the sixteenth century corrosive
sublimate was sold by most druggists. Lemery describes its
preparation by the sublimation of mercuric nitrate (obtained by
evaporating a solution of mercury in nitric acid) with common
salt and calcined green vitriol, but the modern process of manu-
facture was first suggested by Kunckel in 1670. Mercuric sulphate,
obtained by evaporating to dryness a solution -of mercury in hot
concentrated sulphuric acid, is mixed with an equal weight of
common salt. The mixture, to which a little manganese dioxide is
added, is sublimed on a sand-bath in long-necked, flat-bottomed

874 INORGANIC CHEMISTRY CHAP.

flasks : HgS04 + 2NaCl == HgCl2 + Na2S04. The flasks are cooled,
broken, and the cakes of sublimate removed from the upper parts.
Mercuric chloride forms colourless, rhombic needles, sparingly
soluble in cold, but readily in hot water : 100 parts of water
dissolve at 0° 4-3, at 10° 6-57, and at 100° 54 parts of HgCl2. The
salt is only slightly ionised in solution ; less than 1 per cent, is disso-
ciated in decinormal solution, whilst more than 90 per cent, is the
usual ionisation of salts at this dilution. Mercuric chloride is
readily soluble in alcohol and in ether ; if an aqueous solution is
shaken with ether, most of the salt passes into the ethereal layer.

On account of the small ionisation of the salt, mercuric chloride
is not decomposed by boiling concentrated sulphuric acid, but
sublimes unchanged. It is also unacted upon by nitric acid. The
solution of mercuric chloride contains the complex ions HgCl", HgCl3',
and HgCl4". The salt dissolves with evolution of heat in concen-
trated hydrochloric acid ; the resulting solution does not fume,
and on cooling deposits crystals of hydrochloromercuric acid, HHgCl3.
With chlorides of alkali-metals a number of crystalline compounds
are formed, e.g., KHgCl3 and Na2HgCl4, which are partly decom-
posed in solution, and partly ionised into complex ions : Na2HgCl4 ^±
2Na* -f HgCl4". A solution of Na2HgCl4 is used instead of HgCl2
as an antiseptic, since it is neutral in reaction and does not coagulate
proteins.

Alkalies precipitate a solution of mercuric chloride only incom-
pletely, and mercuric oxide readily dissolves in solutions of alkali-
chlorides, forming strongly alkaline liquids : HgCl2 + 2NaOH ^±
HgO -f 2NaCl -f- H2O. This depends on the small ionisation of
mercuric chloride, the concentration of mercuric ions from the
dissociation of which is less than that in the very dilute saturated
solution of mercuric oxide. The latter, therefore, dissolves with
formation of un-ionised chloride.

Mercuric chloride is readily reduced by various reagents, a white
precipitate of calomel or a grey precipitate of metallic mercury
being formed. Stannous chloride produces HgCl or Hg, according
to the amount added. A mixture of mercuric chloride solution
and oxalic acid is reduced (in presence of minute traces of iron salts)
on exposure to light with measurable velocity depending on the
intensity of the light : 2HgCl2 + C2O4H2 = 2HgCl + 2C02 + 2HC1.
Since the calomel may be filtered off and weighed, the reaction
is used as a chemical photometer (Eder).

Phosphorus pentachloride combines with mercuric chloride to form
the volatile crystalline compound, 3HgCl2,2PCl5.

By boiling a solution of mercuric chloride with mercuric oxide, a
series of oxy chlorides is formed, e.g., 2HgCl2,HgO, red ; HgCl2,2HgO
black ; HgCl2,3HgO (kleinite), yellow,

XLI THE METALS OF THE ZINC GROUP 875

Mercuric fluoride, HgF2, unlike the other halogen compounds, is
liydrolysed and forms a basic salt, HgF(OH), with water.
Mercuric bromide, HgBr2, is similar to the chloride.

Mercuric iodide, HgI2. — This salt is formed as a yellow precipitate,
which rapidly becomes scarlet, on adding the calculated amount of
potassium iodide to mercuric chloride solution. On heating to
126°, it is converted into another crystalline form, which is yellow.
The reverse change occurs on cooling, especially if the substance is
rubbed. The yellow form is deposited on sublimation. The
iodide is difficultly soluble in water (1 in 25,000), but readily in
alcohol. It is not decomposed by dilute alkalies.

Mercury periodide, HgI6, is a brown substance obtained by the action
of mercuric chloride on an alcoholic solution of potassium tri-iodide.
It readily loses iodine.

Mercuric iodide readily dissolves in solutions of mercuric chloride
or potassium iodide. In the second case, a complex compound,
potassium mercuri-iodide, K2HgI4, is formed, and can be obtained
as a pale yellow solid on evaporation. The solution is not pre-
cipitated by bases, since practically no mercuric ions are present,
and mercuric oxide dissolves in a solution of potassium iodide to
form a strongly alkaline liquid : HgO -f 4KI -f H20 =K2HgI4 -f
2KOH.

A solution of potassium mercuri-iodide containing excess of
caustic potash is used as a test for ammonia under the name of
Nessler's reagent.

This is prepared by dissolving 62 '5 gm. of potassium iodide in 250 c.c.
of distilled water, and adding to the solution, except 5 c.c. which is
separated, a cold saturated solution of mercuric chloride until a faint
permanent precipitate is formed. About 500 c.c. will be required.
The 5 c.c. of KI solution are then added, and more HgCl2 gradually
until a slight perrnament precipitate is again formed. 150 gm. of
caustic potash are dissolved in 150 c.c. of distilled water and the cooled
solution added gradually to the other solution. The whole is made
up to 1 litre. After settling, the clear solution, which should have a
slight yellow colour, is decanted into a bottle covered with black
varnish. It improves on keeping. With traces of ammonia a brown
colour, with larger amounts a brown precipitate, of NHg2I is formed.

By adding a solution of HgI2 in liquid ammonia to an excess of
potassamide, KNH2, dissolved in liquid ammonia, a chocolate -brown
precipitate of mercuric nitride, Hg3N2, is formed. The acetylide,
3C2Hg,H2O, is formed as a white precipitate on passing acetylene into
a solution of mercuric oxide in aqueous ammonia. The cyanide,
Hg(CN)2, which is only slightly ionised, is formed by dissolving HgO
in aqueous HCN, and crystallising ; it is used in the preparation of

876 INORGANIC CHEMISTRY CHAP.

cyanogen : Hg(CN)2 = Hg + C2N2. The thiocyanate, Hg(CNS)2, is
formed as a white precipitate on adding KCNS to HgCl2 solution ;
when made into small pills and lit with a taper it gives a long, snake-like
mass of mellon, a polymerised cyanogen product (Pharaoh's serpent).

Mercuric carbonate is known only in the form of basic salts ; from
a solution of mercuric nitrate, K2CO3 gives a brown precipitate of
HgCO3,2HgO ; KHCO3 gives a brown precipitate of HgCO3,3HgO.

Mercuric sulphide, HgS. — The sulphide, HgS, which occurs as
cinnabar, is the pigment vermilion. It is formed by triturating
mercury and sulphur with a little caustic potash solution. The
black, amorphous sulphide produced becomes red and crys-
talline on sublimation. Mercuric sulphide is formed by pre-
cipitating a solution of the chloride with sulphuretted hydrogen :
HgCl2 -f H2S = HgS + 2HC1. The black precipitate of HgS first
formed becomes white if shaken with the excess of mercuric chloride
solution, the compound Hg(HgS)2Cl2 being produced. The further
action of H2S changes this into a red and finally a black (HgS)
precipitate. The black precipitate becomes red on sublimation.
It is insoluble in boiling hydrochloric or dilute nitric acid, but
dissolves in aqua regia or in solutions of alkali-sulphides. In the
second case thio-salts, e.g.. K2HgS2,5H20 (white needles), are
formed. The red form of the sulphide is less soluble in alkali
sulphides than the black variety ; the latter when digested with
sodium sulphide solution is slowly converted into scarlet vermilion.
Mercuric sulphide burns when heated in air : HgS + O2 =
Hg + SO2. It is decomposed by heated iron filings : HgS -f- Fe =
FeS -f- Hg (cf. manufacture of mercury).

Mercuric fulminate, Hg(ONC)2, is obtained as a white precipitate
on warming a solution of mercury in excess of nitric acid with
alcohol. It is used in making detonators, since it explodes on per-
cussion. It is now being replaced to some extent by lead azide,
Pb(N3)2.

Mercurammonium compounds. — By the action of ammonia gas
on mercuric chloride, a compound HgCl2,2NH3, called fusible white
precipitate, is obtained. This is also formed as a white precipitate
by adding a solution of mercuric chloride to a boiling solution of
ammonium chloride and ammonia. It was formerly regarded as
mercur-diammonium chloride, Hg(NH3Cl)2, but is probably an
additive compound. If ammonia is added to a solution of mercuric
chloride, mercuric oxide is not obtained, as with potash or soda, but
a white precipitate of mercurammonium chloride, NHg2Cl, i.e.,
ammonium chloride, NH4C1, in which four atoms of hydrogen are
replaced by two atoms of bivalent mercury, is formed. This is
called infusible white precipitate. The brown precipitate obtained
by the action of ammonia on Nessler solution is mercurammonium
iodide, NHg2I.

xi. i THE METALS OF THE ZINC GROUP 877

If mercuric oxide is gently warmed with aqueous ammonia, a
yellow powder known as Millon's base is formed. According to
Rammelsberg (1888), this is the hydroxide corresponding with the
mercurammomum salts, NHg2'OH.2H.2O. On drying at 125° in
ammonia gas, dark-brown explosive NHg2-OH is formed. Hofmann
and Marburg (1899) formulate Millon's base as (HOHg)2NH2-OH.
Compounds isomeric with the salts of Millon's base were pre-
pared by Franklin (1907) by the action of liquid ammonia
on HgBr2 and HgI9 ; he regards them as amino-compounds,
Hg:N-HgX.

By the action of aqueous ammonia on calomel, a black powder
is formed, which is a mixture of infusible white precipitate and
finely-divided mercury, Hg -f- HgNH2Cl. A similar black precipitate
is formed by adding ammonia to a solution of mercurous nitrate,
Hg -f HgNH2-N03. The formation of this black powder from
calomel is said to be the origin of the name of the latter, from the
Greek Jcalomelas, beautiful black. Dry calomel absorbs ammonia
gas, forming the additive compound HgCl,NH3.

If mercuric oxide is dissolved in a solution of potassium nitrite,
and the solution is neutralised with acetic acid, a beautifully
crystallised bright yellow salt, soluble in water, is formed. This is
potassium mercurinitrite, K2Hg(N02)4, and is very stable.

Phosphorescence. — Reference has been made to the phosphor-
escence of calcium sulphide and nitrate, barium and zinc sulphides,
i.e., the property of which these materials possess of shining after
exposure to fight, especially sunlight. This is utilised in the prepara-
tion of himinous paint. Apart from a single unconfirmed observa-
tion, it has always been found that pure compounds do not exhibit
phosphorescence ; the latter is due to traces of heavy metals such
as bismuth, lead, copper, molybdenum, tungsten, uranium, etc.
Thus, phosphorescent calcium sulphide is obtained by heating a
mixture of 100 parts of calcium carbonate with 30 parts of pow-
dered sulphur for an hour to dull redness in a closed crucible. The
mass is cooled, and triturated with alcohol to which sufficient bis-
muth nitrate is added to give 1 part of bismuth to 10,000 of calcium
sulphide. The mass is dried in the air, and heated to dull redness
for two hours. It is then slowly cooled.

Other phosphorescent masses are prepared by heating the mixtures
A below, powdering the product, moistening with the solutions B, and
reheating :

1. Violet light : A : CaO (powder) 20, S 6, starch 2, Na2SO4 0-5,
K2SO4 0-5. B : 2 c.c. of 0-5 per cent. Bi(NO3)3 solution + 0-5 c.c. of
aqueous T12SO4.

2. Deep blue light : A : CaO 20, Ba(OH)2 20, S 6, K2SO4 1, Na2SO4 1,
Li2CO3 2, starch 2. B : 2 c.c. of 0-5 per cent, alcoholic Bi(NO3)3
solution -f- 2 c.c. of 1 per cent. RbNO3 solution.

878 INORGANIC CHEMISTRY CH. XLI

3. Bright green light : A : SrCO3 40, S 6, LioCO3 1, As,S3 1. B : 2 c.c.
of 0-5 per cent. T1NO3 solution.

4. Deep orange-red light : A (only) : BaCO3 40, S 6, Li9CO3 1,
Rb2CO3 0-47.

Lenard explains phosphorescence by supposing that under the
action of light, electrons are emitted from the sulphides of the heavy
metals, e.g., bismuth, but these electrons are retained by the mass of
calcium or barium sulphide. In the dark the electrons return to the
molecules from which they came, and light is emitted when the electron
enters the molecule.

EXERCISES ON CHAPTER XLI

1. Describe the general properties of the sub-group, Be, Mg, Zn, Cd, Hg.
What analogies do beryllium and mercury show to elements of other
groups ?

2. How may a beryllium salt be obtained from beryl ? How has
the atomic weight of beryllium been decided ?

3. In what forms does magnesium occur ? How are Epsom salts
and magnesium chloride made, and for what purposes are they used ?

4. Describe the preparation of : (a) magnesium, (b) anhydrous
magnesium chloride, (c) magnesium nitride. What is the action of
water on these substances ?

5. What are calcined magnesia, magnesia alba* and dolomite ? For
what purposes are they used ?

6. Describe the methods used for the extraction of zinc from its ores,
and for the purification of the metal.

7. How is zinc oxide prepared ? What is the action of (a) dilute
sulphuric acid, (b) ammonia, (c) ammonium sulphide, upon it ?

8. What are lithopone, Rinman's green, turpeth mineral, calomel,
blende, cinnabar, greenockite, fusible white precipitate, mercurius
prcecipitatus per se ?

9. How are the following prepared : zinc carbonate, mercurous
nitrate, mercuric iodide, cadmium sulphide ?

10. How is mercury obtained from its ores, and how is the metal
purified ?

11. How are corrosive sublimate and calomel prepared from mer-
cury ? What is known as to the vapour density of calomel and the
constitution of mercurous salts ?

12. What is the action of ammonia on (a) mercuric chloride, (b)
mercuric oxide, (c) calomel ? What is Nessler's reagent, and what is
its action on ammonia ?

CHAPTER XLII

VOLTAIC CELLS

Electrical energy. — The decomposition of electrolytes by an
electric current is accompanied by an absorption of energy, derived
from the battery or other arrangement used in supplying the
current. In batteries, chemical reactions take place, as a result of
which chemical energy is transformed into electrical energy : this
process may be considered as the inverse of electrolysis. Some
chemists have gone so far as to suppose that all chemical changes
are really cases of reversed electrolysis, but it is evident that the
only changes which can furnish electric currents when carried out
in suitable ways are those involving electrically charged ions, and
there is no reason to suppose that all reactions must occur between
ions.

In a chemical reaction as usually carried out there is generally a
liberation of energy in the form of heat, derived from the^ change of
chemical energy resulting from the atoms falling into new modes of
combination. At first sight it might be supposed that if the
reaction could be carried out so as to produce electrical energy
instead of heat, the former should be equivalent to the latter.
This is not generally the case. The heat evolved in a reaction
which takes place at constant volume is a measure of the change of
total energy in the reacting substances (p. 387). If the reaction is
allowed to take place so as to produce an electric current it is found
that the energy value of the latter may be less than the change of
total energy, in which case the balance is given out as heat, or in
some cases it may be greater than the change of total energy, in
which case the cell absorbs heat from the outside to make up the
balance. The energy of the current is called the free energy of the
reaction, since it may be wholly converted into useful work by means
of an electric motor. The free and total energies of a reaction are
not usually equal.

It was formerly assumed that the heat of reaction, i.e., the
diminution of total energy, was a measure of the work done by the
chemical affinities of the interacting substances, i.e., a measure of
affinity. It is now known that the free energy change is the correct
measure of the affinity. The measurement of changes of free energy

879

880 INORGANIC CHEMISTRY CHAP.

is most conveniently effected by the electrical method, and the
latter is therefore of great importance in chemistry. The question :
" What is the affinity of a reaction ? " is equivalent to the question :
" What is the maximum electrical energy which the given reaction
can yield ? "

Voltage. — Although Faraday's second law shows that the same
quantity of electricity, viz., 96,000 coulombs, is required in the decom-
position of one gram-equivalent of a compound into its uncharged
ions (p. 279), the amounts of electrical work which must be spent in
the decomposition of various compounds are Very different, corre-
sponding with the different affinities. The reason is that the
electrical energy depends on something besides the quantity of
electricity. The decomposition of a gram-molecule each of hydro-
chloric acid and hydriodic acid requires electrical energy equivalent
to 39,300 cal. and 13,100 cal. respectively. Just as the energy of
a stream of water is represented by the product of the volume
flowing past a given section per second and the pressure or head of
water available, so the energy of an electric current is given by the
product of the quantity of electricity transported by the current
and the electrical pressure, which drives the electrons composing the
current. This electrical pressure is called electromotive force, or
voltage. The pressure of water may exist whether the stream is
flowing or not, and the electric pressure may also exist between the
poles of a battery when the latter is not giving any current, and
may be detected by a sensitive electroscope ; it sets the current in
motion as soon as the poles of the battery are joined by a wire
through which the electrons may be driven. The electric pressure
is measured in volts ; this unit is defined in such a way that the
quantity of electricity transported in coulombs, multiplied by
the pressure in volts, gives the electrical energy in joules, where
1 joule = 107 ergs :

Volts X Coulombs = Joules.

The work done per second is equal to the quantity of electricity
moved per second multiplied by the voltage. But the quantity of
electricity moved per second is the current strength in amperes
(p. 282), so that the rate at which work is done by the current, or
the power, is measured by the product of the amperes and volts.
The unit of power, 1 joule per second, is called a volt-ampere,, or a
watt :

Volts X Amperes = Watts.

The watt is a small unit, so that in practice the kilowatt, or 1,000
watts, is used. Energy is then measured in kilowatt hours (K.W.H.),
or the number of kilowatts expended per hour. It is easily seen
that 1 K.W.H. = volts x amperes x 3600/1000.

An ordinary metal -filament lamp uses 220 volts at about half an

XLH VOLTAIC CELLS 881

ampere. The power consumed is 220 x yiOOO = 0-11 K.W., or 110
watts. If the lamp is 220 candle-power, it uses ^ watt per candle, and
is called a " half -watt lamp." Again, 1 cal. =4-186 X 107 erg =
4-186 joules. Thus 1 volt-amp.-sec. = 1 joule = 0-238 cal. The volt-
ages required to decompose hydrochloric and hydrobromic acids are,
on the basis of the numbers given above :

HC1 : 39,300 X 4-186 ~ 96,000 = 1-73 volts ;
HBr 13,100 X 4-186 -f- 96,000 = 0-57 volt ;
since in each case the quantity of electricity involved is 96,000 cmb.

Voltaic cells. — An arrangement in which chemical energy is con-
verted into electrical energy is called a voltaic cell, since the first
representative of this type of apparatus was invented by Volta in
1800. There are many types of such cells, the description of which
belongs to the study of electricity, but one or two representative
forms will be considered so as to make clear the conditions under
which the conversion of chemical into electrical energy takes place.

The earliest type of cell, devised by Volta, consists of a plate of
zinc and one of copper immersed in dilute sulphuric acid. When
the plates outside the liquid are joined by a wire, the zinc dissolves,
but the hydrogen bubbles are evolved from the copper, not from the
zinc. An electric current, recognised by its heating and magnetic
effects, flows through the wire. The direction of flow of positive
electricity is taken conventionally as the direction of the current,
although it is really negative electricity, in the form of electrons,
which flows through conductors (p. 281). With the usual conven-
tion the direction of the current is from the copper to the zinc outside
the cell. Since the current must be completed inside the cell, the
positive electricity passes in the latter from the zinc to the copper.
This is effected by the transport of positive charges by the hydrogen
ions moving in this direction. The hydrogen ions are deposited on
the copper plate, give up their charges to it, and appear as gaseous
hydrogen. The discharge may be regarded as due to the removal
of free electrons from the copper plate, which neutralise the positive
hydrogen ions : H* + € = H.

The negative charge taken from the copper is replaced by a current
of negative electrons flowing along the wire from the zinc to the
copper, i.e., in the opposite direction to the conventional positive
current. These electrons must come from the zinc. The latter
dissolves as positively charged zinc ions, and the positive charges of
the latter are derived by the abstraction of electrons from the zinc
atoms : Zn — 2c = Zn". These electrons remaining in the zinc
pass along the wire to, and neutralise the hydrogen ions arriving at,
the copper plate.

If the zinc had merely dissolved in the acid without producing
current, the hydrogen ions of the acid would have been neutralised

3 L

882 INORGANIC CHEMISTRY CHAP.

in contact with the metal when the latter passed into solution in
the ionic state, and hydrogen gas would have been evolved from
the surface of the zinc. In the cell, the neutralisation of the hydrogen
ions, with production of hydrogen gas, still takes place on account
of the negative charge left by the ionisation of the zinc, but the
hydrogen ions have to travel through the liquid to the copper plate
in order to pick up this charge, so that the two reactions, which
when they take place in the same place give out only heat, when
they are compelled to take place at two different localities produce
a current.

Ordinary zinc contains traces of other metals, such as iron, and the
specks of these metals lying on the surface of the zinc act like copper
plates in the cells. Hydrogen gas is really given off from the second
metal, and the current (which passes round the wire in the cell) flows
through the zinc from the points where solution occurs to the parts
where the impurities lie on the surface. Action of this kind is called
local action. If the surface of the zinc is amalgamated, or if very pure
zinc is used, the impurities are removed, and the surface is uniform.
The metal then no longer evolves hydrogen in dilute acid, since local
action is no longer possible. If, however, the zinc is touched under
the surface of the acid with a piece of copper, or a platinum wire,
bubbles of hydrogen are at once evolved from the wire, and the zinc
dissolves. The copper-zinc couple (p. 182) is really a collection of
little cells, in .which local action takes place. The addition of a little
copper sulphate to the zinc and dilute acid in the preparation of
hydrogen (p. 184) is another instance of local action (cf. tinplate, p. 913,
and galvanised iron, p. 862).

The voltaic cell does not generate electricity. The electrical
charges are present in the chemical substances used in making up
the cell, in the form of electrons, and the electrons are added to, or
subtracted from, atoms to form ions. Some of these ions (e.g.,
hydrogen ions) are discharged in the cell, and other previously
uncharged substances (e.g., zinc) are converted into ions. The elec-
trons leaving one atom and attaching themselves to another are
driven round the outside conducting wire. All the electrons remain
in the materials of the cell, but in new combinations, and none are
set free. During this transfer of electricity, energy may be taken
from the battery. The connecting wire becomes heated, it acts
upon magnets in its vicinity, and if it is cut and the ends are
immersed in an electrolyte, the latter is decomposed. These pro-
cesses involve the expenditure of energy.

The voltage of the Volta cell is about 0-74 ; a large number of
these cells connected in series, i.e., with the zinc of one connected
with the copper of the next, formed the battery used by Davy in
1807 for the decomposition of the alkalies (p. 774).

XLII

VOLTAIC CELLS

883

FIG. 410.— Daniell Cell.

The Daniell cell. — The Volta cell has the disadvantage that its
voltage rapidly decreases when current is taken from it. In another
type of cell, invented by Daniell (1836), the voltage remains prac-
tically constant during action. This
cell consists (Fig. 410) of a rod of
amalgamated zinc immersed in
dilute sulphuric acid, and a plate
of copper immersed in a solution of
copper sulphate. The two solutions
are separated by a pot of unglazed
earthenware, which prevents them
from mixing but permits the
passage of ions moving from one
liquid to the other. The voltage
of this cell is about 1-09.

The action of the Daniell cell is as follows. The zinc dissolves
in the dilute acid as zinc ions, and the copper ions deposit from the
copper sulphate solution as metal. No gas is evolved, since the
hydrogen ions passing from the liquid round the zinc, through the
porous partition, are not deposited but remain in the copper sulphate
solution. Instead of hydrogen ions being deposited on the copper,
copper ions, which are more easily discharged, give up their charges
to, and form a coating of copper on, the copper plate. For every
equivalent of copper deposited, an equivalent of hydrogen ions
enters the copper sulphate solution, forming sulphuric acid, leaving
an equivalent of SO/ ions in the zinc compartment, which form
zinc sulphate with the zinc ions given off by the zinc plate. The
dilute sulphuric acid is therefore gradually converted into a solution
of zinc sulphate, whilst the copper sulphate solution is converted
into dilute sulphuric acid.

The net reaction in the cell is the transfer of two unit positive
charges from the copper ions to metallic zinc,whereby metallic copper
and zinc ions are formed : Cu" -j- Zn = Cu + Zn". Since both ions
are bivalent, the reaction involves the transfer of 2 x 96,000
coulombs, and since the voltage of the cell is !•!, the free energy
of the reaction is 1-1 x 2 x ^6,000 == 211,200 joules, which is
equivalent to 211,200/4-18 = 50,525 gm. cal. The heat evolved
in the displacement of copper from a solution of copper sulphate
by one gm. atom of zinc : Zn -f- CuS04 = ZnS04 + Cu, or :
Zn -f Cu" = Zn" -f Cu, is found experimentally to be 50,100
gm. cal. In this case, therefore, the free energy change is about the
same as, but slightly greater than, the total energy change. This
agreement is exceptional ; in most cells the two quantities are
different ; they may even differ in sign.

Zinc in a solution of zinc sulphate, and copper in a solution of
copper sulphate, separated by a porous partition, will also give a

3 L 2

884 INORGANIC CHEMISTRY CHAP.

current, and may be considered as a modification of the Daniell
cell. In this case the ion SO/ migrates from the CuSO4 to the
ZnS04 solution. Copper is deposited from the first solution, and
zinc dissolves in the second. The former becomes less, and the
latter more, concentrated in the operation of the cell.

If an external voltage slightly greater than 1-1 volts is applied
to the terminals of a Daniell cell in the opposite direction to the
voltage of the cell, the chemical reactions in the latter are reversed.
Zinc is deposited and copper dissolves. This reaction must involve
the absorption of energy by the cell, and since the reversing voltage
need only be infinitesimally greater than the voltage of the cell,
the energy spent in reversing the changes in the cell is, in the limit,
equal to that given out in the direct operation of the cell. A cell of
this type is called a reversible cell. Determinations of chemical
affinity obviously presuppose that the cells are operating rever-
sibly.

Electrode potentials. — If in the Daniell cell the zinc is replaced by
another metal, such as cadmium, the other half of the cell remaining
the same, the voltage changes. This is because the change of free
energy in the new reaction : Cd + Cu" = Cd" -f Cu, is different
from that in the former reaction : Zn -f Cu" = Zn" -f- Cu. If the
zinc half of the cell is retained but silver in silver sulphate solution
is substituted for copper in copper sulphate, there is, for the same
reason, a change in voltage. The voltage of a cell, therefore, depends
on the nature of both its electrodes, i.e., of the metals and solutions.

The voltage also depends on the concentration of the ions in the
solutions around the two electrodes. If the zinc sulphate solution
around the zinc is diluted, or the copper sulphate solution around the
copper made more concentrated, the voltage in each case increases.
If a series of Daniell cells composed of Zn in a solution of ZnS04
containing 1 gm. equiv. of Zn" ions per litre, and other metals in
solutions also containing the unit concentration of metal ions, are
made up, different voltages will result. If a cell composed of two
of these other metals in their solutions, say Cu and Cd, is made up,
its voltage will be found to be the difference between the voltages
of two Daniell cells, composed of Zn and the metals Cu and Cd,
respectively. The voltage of a cell may thus be regarded as the
algebraic difference of two single voltages, one corresponding with
each electrode. These are called electrode potentials.

Electrolytic solution pressure. — The source of the electrode
potentials may be explained by Nernst's theory of electrolytic
solution pressure. A bar of zinc immersed in water, dilute acid, or
a solution of zinc sulphate tends to throw off zinc ions into the.
solution. This tendency is called the solution pressure of the metal.
But the zinc ions in the solution exert an osmotic pressure, and
tend to redeposit on the metal. As a result ot the first change,

XLII

VOLTAIC CELLS

885

Zn -> Zn", the metal will acquire a negative charge, and the solu-
tion containing the zinc ions thrown off, a positive charge. This
reaction is soon brought to a standstill by the attraction of the
opposite charges, so that a layer of positive zinc
ions, which retain their charges, is attracted to
the surface of the negative zinc plate (Fig. 411).

The more zinc ions there are in the solution,
the greater is their tendency to deposit on the
metal, reducing its negative charge, so that the

-Zn-

solution pressure of the metal is opposed and FlG> 411. —Diagram
finally balanced by the osmotic pressure of the ions illustrating Forma-
in solution. The greater the osmotic pressure, Double Layer. n(
the fewer zinc atoms pass out into the solution
as ions, and the smaller is the electrode potential developed. Beyond
a certain concentration of ions, these tend to discharge on the
zinc, and the latter is charged positively.

If a bar of copper is placed in a solution of copper sulphate, the
copper ions of the latter tend, by their high osmotic pressure, which
is opposed by a relatively small solution pressure, to deposit on the
metal, giving up their charges. The metal becomes charged posi-
tively, leaving the solution negatively charged from withdrawal
of positive ions, but the formation of a layer of negative ions on
the surface of the metal again puts a stop to this reaction after a
certain point.

If the two single electrodes, zinc and copper, are put in communi-
cation by a porous partition between the solutions, as in Fig. 412,
we have a Daniell cell. The voltage of this is the algebraic differ-
ence of the single potential differences. The positively charged

copper, the solution pressure
of which is small, tends to
drive a positive current round
from the copper to the zinc
outside the cell, if the metals
are connected by a wire. The
negatively charged, zinc, the
solution pressure of which is
great, tends to drive a negative
current in the opposite direc-
tion to, i.e., a positive current
in the same direction as, the
copper. The ultimate source

-Theorsrof Action of Galvanic of ^ current mfly be regarded

as the superior tendency of the
zinc to force out its ions into the solution.

The single potentials of metals in solutions of their ions containing
1 gm. equiv. per litre are given in the table below. The sign attached

FIG. 412.

Na (+2-4)

+0-16

Ba (+2-6)

H-0-05

Sr -( + 2-5)

—0-02?

Ca (+2-4)

-0-12

Mg +1-3

Sn(Sn")

-0-14

Al +1-03

-0-24

Mn +0-82

-0-53

886 INORGANIC CHEMISTRY CHAP.

to the number of volts is that of the charge of the solution. Thus,
zinc tends, in a normal solution of its ions, to throw out still more
ions until the solution has a positive potential 0-5 volt higher
than that of the metal. Copper ions, on the other hand, will tend
to deposit from a normal solution, leaving the latter negatively
charged at 0-6 volt below the metal. This table is called the
electromotive series. Hydrogen is included, since when dissolved in
platinum or palladium it acts like a metal electrode to solutions of
acids, containing H' ions.

Electromotive series of the metals.

K (+2-6) Fe(Fe'") +0-2 Bi -0-63?

Sb -0-71

Hg(Hg') —0-99

Pd -1-03?

Ag -1-04

Pt —1-10?

Au -1-7?

Zn +0-51 Cu(Cu") —0-58

The voltage of the Daniell cell with normal solutions is therefore
0-51 — (— 0-58) = 1-09, the copper being positive, since the solution
of cojjper sulphate is negative.

Since ionisation takes place by addition of positive charges to the
metal, one metal will dissolve in a solution of another, displacing the
latter, when the electrode potential of the former metal is algebrai-
cally greater. Thus, if a bar of zinc is placed in a solution of copper
sulphate, the zinc tends to throw out ions into the solution. This
tendency, in a normal solution of zinc ions, is measured by 0*5 volt.
Copper ions, on the contrary, tend to deposit from the solution as
metal, since the electrode potential of copper shows that in a normal
solution of its ions the metal becomes positively charged, corre-
sponding with deposition of ions. Zinc in a solution of cad-
mium ions will dissolve, and cadmium will be deposited, since
0-5 — (0*16) = + 0*34 ; whilst cadmium will deposit copper,
since 0-16 — (— 0-58) = -f 0-74. Silver will not deposit copper
from a solution of copper ions, since — 1-04 — (— 0-58) = — 0-46.
These examples show that the electromotive series is an affinity
series.

Although non-metals are non-conductors, their electrode poten-
tials relative to solutions of their ions may be measured by absorbing
a trace of the substance in a platinum plate, and using this as an
electrode. A platinum plate immersed partly in chlorine gas and
partly in a solution containing chloride ions, say IIC1, acts as a
chlorine electrode.

XLII

VOLTAIC CELLS

887

Electromotive series of non-metals.
0-78 Cl -1-50 HSO4

OH -1-96

S04 -2-2

Thus, the voltage of the cell : Zn | NZnSO4 | NKBr I Br0,Pt
will be 0-5 - (- 1-32) = +1-82.

Concentration cells. — Since the electrode potential depends on
the concentration of the ions in the solution, two portions of the
same metal immersed in two solutions of the same salt, of different
concentrations, can form a cell. Cells of this kind are known as
concentration cells. Their voltage obviously cannot depend on
differences of solution pressures, or affinities, since both electrodes
and electrolytes are of the same chemical composition. The
voltage depends on the fact that copper ions in a concentrated
solution of copper sulphate, for instance, tend to
deposit on the copper electrode, on account of
the greater osmotic pressure to a greater extent
than copper ions in a dilute solution of copper
sulphate. The copper plate in the concentrated
solution has a greater positive potential than that
in the dilute solutions, since positive ions are
driven to it with greater force. The metal
dissolves in the dilute solution, and deposits from
the concentrated solution, until both solutions
become equally concentrated. The combination
then shows no voltage.

EXPT. 326. — On a concentrated solution of stan-
nous chloride in a test-tube pour carefully a dilute
solution of the same salt. Insert a stick of tin into
the liquids, holding it by means of a cork, as shown in
Fig. 413. After a few hours a crystalline deposit of tin forms on the
rod in the concentrated solution. In this case the current flows through
the rod from the concentrated to the dilute solution.

If the electrodes are immersed in solutions which are not of normal
concentration with respect to their ions, a correction must be applied
to the electrode potentials given in the tables, to take account of the
influence of ionic concentration. In more concentrated solutions the
osmotic pressures of the ions are more active in tending to cause depo-
sition of the latter on the electrodes. If Pe and P0 are the electrode
potentials of a substance in solutions of its ions of concentrations c and
1 gm. equiv. per litre, respectively, then it can be shown that :

PC = Po -j — log — , where n is the valencv of the ion.

n c

Thus, the electrode potential of Zn in a decinormal solution of its ions

FiG. 413. — Experi-
ment illustrating
action of a Con-
centration Cell.

888 INORGANIC CHEMISTRY CHAP.

is 0-5 + 0-058/2 = 0-53 volt. It is greater than in normal solution,
since the opposing osmotic pressure of the ions is less.

The effect of concentration may be very marked. If a solution
of potassium cyanide is added to the solution of copper sulphate in
the Daniell cell, the copper ions are nearly all removed to form a
complex compound, KCu(CN)2, which ionises as K' and Cu(CN)2',
and the direction of the current actually changes sign. On account
of the low osmotic pressure of Cu" ions, copper dissolves and with
such ease that zinc ions are driven out of solution as metallic zinc.

Colloids- — It has been stated (p. 12) that the particles of colloids are
usually electrically charged. The origin of the electric charge is sup-
posed to be similar to that acquired by metals immersed in liquids,
i.e., ions are given off into the solution, and the particles acquire opposite
charges. The ions are then attracted to the surface of the colloid
particle, and a charged layer is deposited on it. Colloidal metals, for
example, send off a few positive ions into the solution, leaving the par-
ticles with negative charges. Colloidal ferric hydroxide has the
positive charge of the ferric ions which it adsorbs. If water is poured
into a glass vessel, the glass sends off sodium ions, charging the liquid
positively, and the glass acquires a negative charge. If a solution of
copper sulphate is used, sodium ions pass into it, and copper ions are
adsorbed by the negative glass surface. If the glass vessel is washed
with water, the copper is not removed, but it is dissolved off by acid.
A very dilute solution of copper sulphate, which is toxic to sprouted
pea-seedlings, is rendered non-toxic by shaking with powdered glass,
since the latter adsorbs the copper ions. The charge on colloidal
arsenious sulphide is derived from the sulphuretted hydrogen used
in its preparation :

(As2S3)M + H2S = H2S(As2S3)w = H" + HS(As2S3)n'.

Oxidation and reduction. — The oxidation of stannous chloride to
stannic chloride, or ferrous chloride to ferric chloride, by means of
chlorine, may be made to furnish an electric current. A cell is
made up as follows :

Positive pole : a platinum plate in a solution of a chloride, say NaCl,
saturated with chlorine.

Negative pole : a platinum plate immersed in a solution of stannous
chloride.

The two are separated by a porous partition. Chlorine dissolves
in the platinum, and sends off chloride ions into the solution. The
plate is thus left with a positive charge. To neutralise this, nega-
tive electrons pass round the wire from the other plate, and the
stannous ions which come in contact with this lose negative charges
and are oxidised to stannic ions : Sir" — 2e = Sn'" , The current

XLII VOLTAIC CELLS 889

is completed in the cell by chloride ions moving through the porous
partition.

Let a cell be constructed as follows :

Negative pole : a platinum plate charged with hydrogen immersed in
dilute acid.

Positive pole : a platinum plate in a solution of ferric chloride.

The following action occurs. Hydrogen dissolved in the negative
plate throws off hydrogen ions into the solution, leaving the plate
charged negatively. The negative charge passes to the other plate,
and discharges any Fe"' ions touching the plate to Fe" ions. This
is a process of reduction.

By measuring the voltages of cells of this kind, one can determine
the relative strengths of oxidising and reducing agents. The follow-
ing table gives the potentials of oxidising and reducing reagents ;
the sign of the potential is that of the solution, as before. The
electrodes are platinum plates.

Alkaline SnCl2 +0-30 KI -0-89

Alkaline NH2- OH +0-06 K3FeCy6 -0-98

H2 -0-25 K2Cr2O7 -1-06

NaHSO3 -0-66 KNO2 -1-14

AcidFeSO4 -0-78 KMnO4 -1-76

EXERCISES ON CHAPTER XLII

1. Describe with examples how chemical energy may be converted
into electrical energy. How is the reverse change effected ?

2. In what units is electrical energy measured ? What are the units
of voltage and current, and how are they connected with the unit of
energy ?

3. What is meant by total energy and free energy ? How are they
connected with the chemical energy change in a reaction ?

4. The voltage of a Weston cell (p. 87l) is 1-0186 volts. Find the
free energy of the reaction taking place in the cell.

5. What is meant by single potentials and electrolytic solution
pressure ? Give an account of the action of the Daniel! cell on the
basis of Nernst's theory of electrolytic solution pressure.

6. Calculate the voltage of a cell composed of a platinum plate
saturated with hydrogen immersed in normal acid as one electrode and
a platinum plate in contact with solid iodine in a normal solution of
potassium iodide as the other electrode (both electrolytes may be taken
as 90 per cent, ionised). Find the free energy of the reaction in the
cell, and compare with the heat of formation of hydrogen iodide (p. 410)
in solution.

7. How may the oxidising strengths of reagents be compared?

CHAPTER XLIII

THE METALS OF GROUP HI OF THE PERIODIC SYSTEM

Group III of the periodic system. — Group III in the periodic
table is divided into two parts :

The Boron Sub-group : Boron and the Metals of the Rare Earths.

The Aluminium Sub-group : Aluminium, Gallium, Indium, and
Thallium.

All these elements form oxides, R203, and chlorides, RC13. They
are generally tervalent. Thallium, however, forms univalent, T1X,
as well as tervalent, compounds. Boron trioxide is a weakly acidic
oxide, but shows feebly basic properties towards very strong acids.
The basic character of the oxides increases with the atomic weight.
The hydroxides of the aluminium sub-group are amphoteric3 forming
salts both with strong acids and with strong bases :

A1(OH)8 + 3HC1 = A1C18 +3H2O; and
A1(OH)3 + NaOH = NaA102 + 2H20.

The metals of the aluminium sub-group do not oxidise very easily
in the air, though this tendency increases with the atomic weight.
Aluminium is fairly stable in the air, whilst thallium oxidises
moderately easily. Aluminium, gallium, and indium form charac-

i ni

teristic alums with the formula : M2S04,R2(S04)3,24H20, which
are isomorphous, M being a univalent metal. A true thallium alum
has not been prepared.

The element boron, a non-metal, has already been described. The
rest of the elements of the group, with the exception of aluminium,
are rare, so that they will be described briefly after aluminium.

ALUMINIUM. Al = 26-8.

Aluminium. — Aluminium is the most widely distributed light
metal on the surface of the earth. It occurs to the extent of 7*3
per cent, in the earth's crust, as silicates in almost all crystalline
silicate rocks (felspar, augite, hornblende, tourmaline, and micas),
and in the secondary formations day (Al203,2Si02,2H20), and slate
(clay hardened and laminated by pressure). The oxide is found,
either anhydrous, as corundum, A1203, or hydrated as diaspore,

890

CH. XLIII METALS OF GROUP III OF THE PERIODIC SYSTEM 891

A1203,H9O. bauxite, A1203,2H2O (amorphous), and hydrargillite,
A1203,3H2O. Felspar, KAlSi3O8, or K2O;Al2O3,6SiO2, is a con-
stituent of primary rocks such as granite, and by the disintegration
of these rocks, either by simple hydrolysis or by the combined
action of moisture and atmospheric carbon dioxide, soluble alkali
salts and insoluble hydrated aluminium silicates (clays) pass into
the soil :

2KAlSi3O8 -f 3H20 = Al203,2Si02,2H2O + 4Si02 + 2KOH.
2KAlSi308 + 2H2O + C02 = Al2O3J2Si02,2H2O + 4Si02 + K2C03.

The quartz crystals and mica scales of such primary rocks as
granite remain in situ along with the fine deposit of clay, or kaolin,
Al203,2SiO2,2H20, derived from the felspar. Any iron present
in the rock is oxidised to ferric oxide, which colours the clay yellow
or red. The kaolin may be separated from the quartz by washing,
when the fine particles of clay are carried away from the larger
pieces of quartz. Fine particles of pure clay are separated from a
slightly alkaline suspension by cataphoresis (p. 12).

Common clay is contaminated with limestone, quartz, and oxide
of iron ; a mixture of clay and limestone constitutes marl, whilst a
mixture of clay and sand is called loam. Nearly all clays contain
small amounts of titanium oxide, Ti02. Aluminium compounds
are not absorbed (except in traces) from soils by plants, with the
exception of mosses.

Other aluminium minerals are spinel, MgAl204, and chrysoberyl,
BeAl204, in which alumina plays the part of an acidic oxide.
Cryolite. Na3AlF6, is a semi-transparent rock, found in large masses
in Greenland. The turquoise is a basic aluminium phosphate,
coloured blue or green with copper phosphate.

Alum, from which the element takes its name, was probably
known to the ancients ; Paracelsus observed that it was not a vitriol
(i.e., a compound with a metallic basis), and Pott (1746) showed that
it was derived from a peculi . earth, alumina, which Marggraf
(1754) was first able to isolate from clay. That this earth was the
oxide of a metal was regarded as certain by Davy, but the metal,
aluminium itself, was first isolated by Wohler in 1827 by the action
of sodium on the chloride, A1C13. Bunsen (1354) prepared alumi-
nium by the electrolysis of the chloride, but the first industrial
method of preparation, due to Deville (1854), depended on the
reaction used by Wohler. In 1886 the manufacture of aluminium
by the electrolysis of alumina dissolved in fused cryolite was started
simultaneously by Hall in America and by Heroult in Europe,
where the two processes, differing only in detail, are now used on an
extensive scale.

On account of the small chemical equivalent of aluminium (9), and
the very high heat of formation of the oxide : 2A1 -f 3O = A12O3 + 380

892

INORGANIC CHEMISTRY

CHAP.

kg. cal., a large expenditure of energy is required, which can be obtained
economically only from cheap water power.

The reaction is carried out in the electric furnace. The Hall process
is worked by the Aluminium Company of America, utilising water
power at Niagara, Massena, and Shawinigan Falls. The ' Heroult
process is applied by the Societe Electrometallurgique Francaise, at
Froges, and by the British Aluminium Company at Kinlochleven in
Scotland.

Manufacture of aluminium. — It has not yet been found possible
to produce aluminium from clay ; the source of the metal is bauxite,
but since this contains iron it is first treated to obtain pure alumina,
A1203.

In Germany, the bauxite is heated to bright redness with sodium
carbonate, when sodium aluminate, NaA102 or Na20,Al203, is pro-
duced, alumina being a feebly acidic oxide. The mass is rapidly

lixiviated, forming finely-
divided oxide of iron,
which can be used for
the purification of coal
gas (p. 682), and a
solution of sodium
aluminate, from which a
granular precipitate of
aluminium hydroxide,

A1(OH)3 or A12O3,3H20,
is thrown down by
carbon dioxide at

FIG. 414. — Electric Furnace for Aluminium : A.
Carbon Anodes ; B. Carbon Lining ; C. Cast-iron
Vessel ; D. Carbon Powder Protection ; E. Crust of
Solidified Electrolyte; F. Molten Electrolyte; G.
Molten Metal ; H. Low Voltage Charge Control Lamp.

50-60° : 2NaAlO2 + C02 + 3H20 = Na2CO3 + 2A1(OH)3. On
igniting the precipitate, A12O3 is obtained, and the solution of
Na2C03 is evaporated and used again. The British Aluminium
Co., at Larne (Ireland), uses the Bayer process. The bauxite is
digested in kiers with caustic soda solution under 80 Ib. pressure,
giving a solution of sodium aluminate, and leaving oxide of iron,
which, however, cannot be used for any purpose. The solution of
sodium aluminate is now digested with precipitated alumina, when
nearly all the alumina in solution is thrown out as a sandy, amor-
phous precipitate (/J-Al^Og), which is easily washed, and on
ignition yields pure alumina.

The electric furnace consists of an iron box, 6 ft. by 3 ft. by 3 ft.,
lined with blocks of carbon, which is made the cathode. The
anodes consist of rods of petroleum coke or gas-carbon set in a row
(Fig. 414) about 2-3 in. above the bottom of the trough. The electro-
lyte is a solution of alumina (m.-pt. 2010-2050°) in fused cryolite,
together with some fluorspar, the temperature being kept at
875-950°.

XLTII METALS OF GROUP III OF THE PERIODIC SYSTEM 893

The eutectic point for a mixture of A12O3, cryolite, and CaF2 is 868°,
and occurs when these are in the proportion 17-7 : 59-3 : 23. In prac-
tice, the mixture used is A12O3 10-25, cryolite 90-75, fluorspar 36 per
cent, of the cryolite.

The alumina only is electrolysed, the metal (m.-pt. 659°) forming
a pool below the anodes, and the oxygen burning the anodes to CO.
An E.M.F. of 5-6 volts, and an anodic current density of 100 amp.
per sq. dcm., corresponding with a total current of 10,000 amp.,
are used. The charge is covered with a layer of carbon, and fresh
alumina is stirred in from timfe to time to replace that decomposed.
The decomposition is indicated by a rise in resistance, the shunted
lamp brightening. About 165,000 tons of aluminium are produced
annually in America, France, Great Britain, Canada, Switzerland,
Austria, and Germany. The metal has a purity of 99 per cent. Al,
the impurities being chiefly iron and silicon.

Properties of aluminium. — Aluminium is a white metal with a
blue tinge, the density varying from 2-703 (cast) to 2-709 (rolled).
On account of its very small density it has been used in the con-
struction of airships, and engine parts, etc., of motor-cars ; the
alloy magnaUum (90-98 Al and 10-2 Mg) is still lighter, and can be
worked easily in a lathe, whilst duralumin (94-4 Al + 0-95 Mg +
4-5 Cu + 0-76 Mn), sp. gr. 2-77-2-88, can be worked hot or cold, and
hardened by quenching from 250-520° in water, the hardness
being increased by tempering up to the melting point (520°)
after quenching. Duralumin is Used in airship construction. Alloys
of aluminium with copper are called aluminium bronzes (e.g., 90 Cu
+ 10 Al).

Aluminium melts at 658-5°, and boils at 1800°. Its tensile
strength is high : cast 7, sheet 11, wire 13-29 tons per sq. in., that
of mild steel being 25. Its surface is unaltered in air, a thin,
transparent protecting film of oxide being formed. If this film is
removed by rubbing the metal with mercuric chloride, producing
a liquid amalgam to which the oxide cannot adhere, rapid oxidation
with production of moss -like excrescences of oxide occurs. Amal-
gamated aluminium foil is also a useful reducing agent, since it
reacts in neutral solutions. Aluminium foil or powder readily
burns in air with a brilliant flame when heated.

The metal can be cast ; at 100-150° it can be wrought, rolled,
or drawn, but it becomes brittle at 600°. It is a good conductor
of heat and electricity, being used for electric cables in America.

Aluminium can be soldered, but only if a special solder
(Al 2-25 -4- phosphor-tin 0-75 -+- zinc 17 + tin 80) is used, which is
first applied to the metal by heating to 600 °, and the two surfaces
then pressed together.

About one-thousandth of the weight of aluminium added to
molten steel before casting removes oxygen and nitrogen, forming

894 INORGANIC CHEMISTRY CHAP.

A12O3 and A1N, and prevents blow-holes in castings. It reacts
very violently with silicon steels.

Although only superficially attacked by pure water, aluminium
is strongly attacked by sea-water or saline solutions, holes being
rapidly formed. Dilute sulphuric acid has very little action on
aluminium and the pure metal is almost unattacked by dilute or
concentrated nitric acid. Dilute and concentrated hydrochloric
acids readilv dissolve the metal with evolution of hydrogen :
2A1 + 6HC1 = 2A1C13 + 6H2. Concentrated sulphuric acid attacl
aluminium only when heated :

2A1 + 6H2S04 = A12(S04)3 + 3S02 + 6H20.

The metal readily dissolves in solutions of alkalies, forming alumin-
ates : 2A1 + 2NaOH + 2H90 - 2NaA102 + 3H2. These are
hydrolysed in solution : NaAl02 + 2H2O =± A1(OH)8 + NaOH.

The great evolution of heat resulting from the combination of
aluminium with oxygen is utilised in Goldschmidt's thermit process
for reducing metallic oxides (e.g., Cr203, Mn02); and for the pro-
duction of molten steel for welding broken articles (rails, ships'
propellers, etc.) in situ. A mixture of aluminium powder and
oxide of iron (" smithy-scales ") is placed in a crucible, and ignited
by a magnesium wire (p. 948). A violent reaction occurs :
2A1 -f Fe2O3 = 2Fe -f- A12O3, and molten iron, covered with a
layer of molten alumina, is formed. The iron is tapped from below
directly on to the joint to be welded.

If two plates of aluminium are immersed in sodium bicarbonate
solution, and connected with an alternating current supply, the latter
is converted into a pulsating direct current. The film of oxide on the
metal offers a very high resistance to the current when the plate becomes
an anode ( + )» whilst the metal as a cathode ( — ) allows the current to
pass freely. This arrangement is known as an aluminium rectifier.
They have been replaced to some extent by thermionic valves (i.e.,
vacuum tubes with electrodes emitting electrons on heating).

Alumina, A1203. — Aluminium trioxide, or alumina, A1203, is the
only oxide of aluminium known with certainty. It occurs native
as corundum, which forms rhombohedral crystals nearly as hard
as the diamond ; emery is an impure fine-grained variety, used in
grinding and polishing. Corundum, when transparent, forms a
number of gems : oriental topaz (yellow) ; sapphire (blue, due to
Co, Cr, or Ti oxides) ; ruby (red, due to Cr203) ; oriental amethyst
(violet, due to Mn) ; oriental emerald (green).

Artificial rubies are produced (Verneuil, 1904) by dropping powdered
alumina containing 2-5 per cent, of chromium sesquioxide through the
centre of an oxy hydrogen flame. The fused mass, or " boule," is caught
on a rod of alumina ; it is not amorphous, but forms a single crystal,

XLIII METALS OF GROUP III OF THE PERIODIC SYSTEM 895

which may be cut. Artificial sapphires are made with alumina to
which 1-5 per cent, of Fe3O4 and 0'5 per cento of TiO2 are added ; a
reducing flame is used.

Ahindum, used as a refractory, is prepared by fusing bauxite in
the arc furnace at 2050°, allowing the impurities to settle, cooling,
and crushing the upper part. The powder is mixed with a little
clay and felspar, moulded, dried, and fired in a porcelain kiln at
1500°. It differs from silica in being a basic refractory.

If an alkali is added to a solution of an aluminium salt, e.g., alum,
a white, gelatinous precipitate of aluminium hydroxide, A1(OH)3,
is produced, soluble in excess of potash or soda, but insoluble in
ammonia. If this is dried in the air it has the composition :
A1(OH)3, or A1203,3H2O. When dried at 80°, it has approximately
the composition A12O,,2H20. The sandy powder precipitated in
Bayer's process (p. 892) has the composition A1203,3H2O. At
225°, it forms A1203,2H20, occurring naturally as bauxite (but
usually with less water), and at 235° it gives A1203,H20.
A1203,H20 is also said to be produced by precipitating a boiling
solution of an aluminium salt with ammonia, and drying at 100°.
If these hydrates are heated to dull redness, alumina, A1203, is
left as a white powder. Alumina, when calcined at a low tem-
perature, is soluble in acids, but if strongly heated it becomes
denser (2-8 at 600°, 3-9 at 1200°), and insoluble in acids. The
change appears to take place at 850°, and the product can then be
brought into solution only by fusion with caustic soda or potassium
bisulphate. In the first case an aluminate, in the second aluminium
sulphate, is formed.

Apparently some kind of polymerisation occurs on heating, and the
change is probably exothermic. Other oxides, e.g., O2O3,Fe2Oa,
MgO,TiO2, undergo more or less sudden exothermic changes at higher
temperatures, increasing in density and becoming insoluble in acids,
and generally less reactive. In the case of zirconia, ZrO2, especially,
but to a less extent with some of the other oxides, the change is accom
panied by incandescence. These changes have been little investigated,
since their observation by Berzelius.

Precipitated aluminium hydroxide readily carries down by
adsorption various colouring matters and colloidal substances.
Hence alum and aluminium salts are largely used as mordants in
dyeing, and for clarifying , water and liquids such as sewage, the
calcium carbonate dissolved in which precipitates alumina. In
mordanting, the alumina is first precipitated in the fabric, and the
latter dipped into the solution of the dye. In clarification, lime-
water is added to precipitate alumina.

EXPT. 327. — Take two pieces of clean white cotton cloth. Dip one
into a solution of aluminium acetate, and dry on the water-bath.

;

896 INORGANIC CHEMISTRY CHAP.

Immerse the two pieces in two beakers, containing boiling solutions of
logwood extract ; take out after ten minutes and place in beakers of
boiling water. The colour is retained by the mordanted fabric, but is
leached out of the other. The adsorption product is called a lake.

Fabrics are also waterproofed by steeping .in a solution of alumin
ium acetate (q.v.), and steaming, when colloidal alumina is pre-
cipitated in the pores of the fabric (" rainproof s ").

Colloidal aluminium hydroxide exists in two forms, (a) The pre-
cipitated hydroxide is soluble in a solution of aluminium chloride,
and the solution on dialysis yields a colloidal aluminium hydroxide,
which acts as a mordant, and is coagulated by alkalies or salts,
the precipitate being soluble in acids (Graham, 1861). (b) If a
solution of aluminium acetate is kept for some time at 100° in an
open flask, the water which evaporates being replaced, all the acid
is expelled, and a second colloidal variety (meta-aluminium hydroxide)
is formed, which does not act as a mordant ; it is precipitated by
acids, alkalies, and salts, but the gel is sparingly soluble in acids
(Crum, 1854). The gel, dried at 100°, has in each case the com-
position A12O(OH)4. A milky colloidal solution is also formed
by the action of 4 per cent, acetic acid on the well-washed pre-
cipitated hydroxide.

Aluminium peroxide, A12O4, is precipitated, mixed with alumina,
by adding excess of 30 per cent. H2O2 to alumina dissolved in 30 per
cent, caustic potash solution.

Aluminates. — Aluminium hydroxide readily dissolves in
acids, producing aluminium salts, and thus acting as a base :
A1(OH)3 + 3HC1 ;=± A1C13 + 3H20. The reaction is reversible, and
the salts are hydrolysed by water, indicating that . aluminium
hydroxide is a weak base. The hydroxide also dissolves in solutions
of alkalies, producing aluminates, which are extensively hydrolysed
by water ; it is therefore capable of acting also as a weak acid.
The acidic properties are weaker than the basic ; they are caused
by the hydroxyl groups splitting off hydrogen ions. This goes
on only in two stages ; the normal aluminates, e.g., Na3A103,
apparently do not exist.

AT-' + 30H' — A1(OH)3 ^± H' + H2A103' ^± H' + A102' + H20.

In solution, only the meta-aluminates, RA102, appear to exist,
since the freezing point of a solution of caustic soda is unaltered
by dissolved alumina, so that an OH' ion is replaced by A102' :
OH' + A1(OH)3 = A1O2' -f 2H20. If solutions of equal amounts
of alumina in acid and alkali, respectively, are mixed, the whole
of the alumina is precipitated : AT" + 3A1O2' = 2A1203. Solu-
tions of aluminates are so largely hydrolysed : NaA102 + 2H,,0 ^
NaOH + A1(OH)8 ^± Na' + OH'' + A1(OH)8, that they may be

XLIII METALS OF GROUP III OF THE PERIODIC SYSTEM

S97

titrated with acids as if they were caustic alkalies, and on standing the
alumina is slowly deposited. They do not appear to contain col-
loidal alumina, the slow deposition corresponding with a slow
hydrolytic change. When boiled with alumina, all the aluminium
hydroxide is precipitated.

Various aluminates occur as minerals, e.g., spinel, MgAl204 or
MgO,Al.2O3. The Mg may be replaced isomorphously by Fe (ous),
Mn (ous), Zn, etc., and the Al by Fe (ic), Cr (ic), Mn (ic), etc. — all
the minerals being classed as spinels. Cobalt aluminate, CoAl204,
is formed as a blue mass on igniting alumina with cobalt nitrate
(blowpipe test for Al), and is known as Thenard's blue.

Halogen compounds of aluminium. — Anhydrous aluminium
chloride, A1C13, is formed by heating aluminium in hydrogen chloride :

FIG. 415. — Preparation of Aluminium Chloride.

2A1 -f 6HC1 = 2A1C13 -f 6H2, or by heating strongly a mixture of
alumina and carbon in a current of chlorine :

|3C12 + Ayp3+3C= 2A1C13 -f 3CO.

Alumina is not decomposed by chlorine, or by carbon alone below
2000° ; the combined affinities of chlorine for aluminium, and of carbon
for oxygen, however, bring about the decomposition.

EXPT. 328. — Heat 10 gm. of aluminium turnings in a hard-glass tube
connected with a bottle (Fig. 415) and pass over the metal a current of
hydrogen chloride dried by sulphuric acid. A sublimate of aluminium
chloride is formed, which may be collected in the bottle by heating the
tube. When the whole has passed into the bottle, fit a good cork to
the latter, as the substance is very hygroscopic.

Anhydrous aluminium chloride is a white, crystalline substance
(usually coloured yellow by ferric chloride as impurity), subliming
at 183° without previous fusion (m.-pt. 193° under 2 atm. pressure).

3 M

898 INORGANIC CHEMISTRY CHAP.

The vapour density at 183° corresponds approximately with the
formula A12C16, but rapidly diminishes with rise of temperature,
until at 450° it corresponds with A1C13, remaining constant at
higher temperatures : A12C16 ^± 2A1C13. In organic solvents, the
formula is A1C13 ; a compound with nitrobenzene has the formula
A12C16,C6H4N02 in solution in carbon disulphide.

Aluminium chloride fumes in the air, and is very deliquescent.
With a little water it forms a crystalline hydrate, A1C13,6H2O, which
is more conveniently prepared by dissolving aluminium, or alumina,
in concentrated hydrochloric acid, and saturating the solution with
hydrogen chloride gas. It is hydrolysed in solution : A1C13 -f 3H2O ^±
A1(OH)3 -f- 3HC1 ; the latter has an acid reaction, and can be titrated
with alkali as if it were free hydrochloric acid. The anhydrous
chloride forms the compounds A1C13,6NH3, A1C13,SC14 ; double salts,
e.g., NaAlCl4, are formed by crystallising a mixed solution of the
chlorides.

Aluminium bromide, AlBr3, and iodide, A1I3, are formed by passing
HBr or HI over heated aluminium. Their properties are as follows :

AlBr3 : m.-pt. 93°, b.-pt. 263° ; vapour density Al2Br6 ; in solution
in CS2, Al2Br6 ; in nitrobenzene, AlBr3. Forms a crystalline hydrate,
AlBr3,6H2O.

A1I3: m.-pt. 125°, b.-pt. 350°; vapour density, A12I6 ; in solution
A12I6. Forms crystalline hydrate, A1I3,6H2O. Reacts with carbon
tetrachloride to form CI4:4A1I3 + 3CC14 = 4A1C13 + 3CI4.

Aluminium fluoride, A1F3, is formed similarly to the chloride,
but is much less volatile, and is scarcely soluble in water. Although
alumina dissolves in hydrofluoric acid, the solution is strongly
supersaturated, and soon deposits the fluoride. Seven hydrates
are described. The salt dissolves in hydrofluoric acid, probably
forming hydrofluoaluminic acid, H3A1F6, a salt of which is cwjolite,
NagAlFg, which may contain tervalent fluorine : Al:(F:FNa)3.

Cryolite is used as a flux in the manufacture of aluminium. It
has also been used as a source of soda and alumina by Thomson's
process. Powdered cryolite (separated from gangue, etc., by
electromagnetic processes) is heated with lime : Na3AlF6 -f- 3CaO =
3CaF2 -j- Na3A103. The aluminate is dissolved out, and decom-
posed by carbon dioxide :

2Na3A103 + 3H2C03 = 3Na2C03 + 2A1(OH)8.

Aluminium sulphate, A12(S04)3S — If alumina is dissolved in hot
concentrated sulphuric acid, the liquid on cooling slowly
deposits an indistinctly crystalline mass of aluminium
sulphate, A12(S04)3,18H20. This is purified by redissolving in a
little water and adding alcohol. An oily supersaturated solution
separates, which soon solidifies to lustrous, scaly crystals of the

METALS OF GROUP III. OF THE PERIODIC SYSTEM 899

above formula. On heating the crystals they intumesce, leaving
a white mass of anhydrous sulphate, A12(S04)3. Many other
hydrates have been described.

Impure aluminium sulphate is made by heating kaolin
(clay) with concentrated sulphuric acid, or bauxite with dilute
sulphuric acid. In the first case silica separates :

Ala08,2Si02,2H20 -f 3H2S04 = A12(SO4)3 + 2Si02 + 5H2O ;

the mass is run into moulds, and solidifies. In the second case,
the settled solution is evaporated, and the crystals are pressed.
The product may contain a considerable amount of ferric sulphate
(especially if bauxite is used) which, although it does not form
mixed crystals with aluminium sulphate, cannot be separated from
it by crystallisation. If the ferric is reduced to a ferrous salt,
say by sulphuretted hydrogen, the aluminium sulphate may then
be crystallised out alone. The crude mixture, known as alumino-
ferric, is used for the precipitation of colloidal matter from sewage
(p. 895).

If precipitated aluminium hydroxide is dissolved in a solution of
aluminium sulphate, a basic salt is deposited : Al203,2S03,o:H2O.
The salt A12(OH)4S04,7H20, or A1203,S03,9H2O, occurs as
webster ite, used in the preparation of alum.

Alums. — The name alum was given originally to a double salt
of aluminium sulphate arid ammonium sulphate,

(NH4)2S04,A12(S04)3,24H20,

which readily crystallises in octahedra. It was prepared from
alum shale, i.e., aluminium silicate permeated by pyrites, FeS2,
which on roasting in heaps forms a mixture of aluminium and
ferric sulphates. The roasted shale is lixiviated, and after evapora-
tion, either ammonium sulphate (originally ammonium carbonate,
i.e., stale urine), or potassium sulphate or chloride, added. The alum
is deposited. Potash alum is prepared from alunite, or alum-stone,
K2SO4,Al2(S04)3,4Al(pH)3, by heating to 500-600°, digesting with
concentrated sulphuric acid, and adding potassium sulphate. Alum
is now usually made by adding the alkali-sulphate to a solution of
alumino-ferric. Since alum is readily purified by recrystallisation,
it may be obtained free from iron (which gives dull colours to lakes
in mordanting) very much more readily than aluminium sulphate.
Alum prepared from alunite, called Roman alum, is quite free from

iron.

If caustic potash is added to a solution of alum, the precipitate of
alumina at first redissolves on stirring, but at a certain point a permanent
precipitate begins to form. The crystals deposited from this solution
on heating to 40° are known as neutral alum and are identical in com-
position with alunite. They redissolve on cooling. If a little alkali is

3 M 2

900 INORGANIC CHEMISTRY CHAP.

added to a solution of alum, the latter, on evaporation, separates in
cubes. Potash-alum appears to effloresce in air ; in reality ammonia
is absorbed from the atmosphere, and a basic salt is formed.

Potash-alum, K2SO4,A12(S04)3,24H2O, when heated melts at
92°, and loses the whole of its water at 200°, forming a white,
porous mass of burnt alum. Ammonia-alum on the other hand,
which melts at 95°, loses ammonia and sulphuric acid as well, and
on ignition leaves a residue of pure alumina :

(NH4)2S04,A12(S04)3,24H20 = 2NH3 + 4H2SO4 + A1203 + 21H20.

The name alum is given to all double-salts of the type
i in
R2S04,M2(S04)3,24H20.

I i

R may be K, NH4, Na, Cs, Tl, hydroxylamine, or the radical of
an organic quaternary nitrogen base, such as N(CH3)4. (Li gives no

in in in in in in in in in in

alum.) M may be Al, Fe, Cr, Mn, In, Tl, Ga, V, Co, Ti, Mn, Rh, etc.
The radical SeO4 of selenates may replace S04. An alum containing
uni- and ter-valent thallium together does not exist, although
i in

T12S04,A12(S04)3,24H2O exists, and ammonium alum containing
thallic sulphate, in mixed crystals, is known. All the alums are
isomorphou.s, form mixed crystals in all proportions, and also
" layer-crystals," i.e., a crystal of any one alum continues to grow
in a solution of any other. The sodium alum is very soluble, and
its preparation is difficult.

Aluminium sulphide, A12S3, is formed from its elements, or by
passing sulphur vapour over a heated mixture of alumina and
carbon. It is completely hydrolysed by water : A12S3 '-{- 3H2O =
2A1(OH)3 -j- 3H2S, and is not formed by adding ammonium sulphide
to a salt of aluminium ; in this case aluminium hydroxide is pre-
cipitated, and sulphuretted hydrogen evolved : Al'" -f- 3HS' +
3HaO = A1(OH)3 -f 3H2S (cf. Cr, p. 953).

Aluminium nitride. — Aluminium combines directly with nitrogen
at 740 °, forming the nitride, A1N, in small yellow crystals, or as a
bluish-green powder. The impure nitride is formed by heating a
mixture of bauxite and carbon to 1600° in a current of nitrogen :
2A12O3 + 6C + 2N2 = 4A1N -j- 6CO. At 1850°, the nitride decom-
poses. When the impure nitride is heated in a carbon tube at 2020°
in a stream of nitrogen, colourless hexagonal needles of pure nitride
are formed. Aluminium nitride is decomposed bv hot dilute alkali,
with evolution of ammonia : 2A1N -f 3H20 = A1203 + 2NH3.
This is the Serpek process, formerly used for the fixation of atmo-
spheric nitrogen.

Aluminium nitrate, A1(N03)3,18H2O, is prepared by mixing solutions

XLIII METALS OF GROUP III OF THE PERIODIC SYSTEM 901

of aluminium sulphate and lead nitrate, filtering, and evaporating.
Other crystalline hydrates (15, 16, or 12H2O) are known. A solution
of the salt is used as a mordant. Aluminium acetate, A1(C2H3O2)3, is
obtained from lead acetate and aluminium sulphate.

Ceramics.— The manufacture of porcelain, carried out by the
early Chinese and Egyptians, remained a lost art in Europe until
1709. when it was rediscovered by Bottcher ; in 1710 the famous
Meissen works in Saxony was started. After 30,000 experiments,
Pott, in Prussia, also rediscovered the secret, and the Berlin works
was begun. The process was rediscovered in France, chiefly owing
to the work of Reaumur, about 1758, and the Sevres works was
established in 1767. The earlier work of Bernard Palissy (1509-
1589) was directed mainly to the glazing and colouring of pottery,
or earthenware, as distinct from porcelain.

The production of pottery (which was carried to a high stage of
perfection by the Etruscans) and of porcelain depends on the
changes produced in clay by heating (or " firing "). Pure clay
(kaolinite) has the formula Al203,2SiO2,2H20. On heating, moisture
is first driven off and colloidal matter coagulated. At 500°, the
kaolinite decomposes : Al2O3,2SiO0,2H2O - A1808 -f 2SiO2 + 2H2O
(or Al2O3,2SiO2 + 2H20) ; at 800°, the alumina begins to poly-
merise, and the mass shrinks ; above 1000°. combination occurs
between alumina and silica, with formation of sillimanite, Al2O35SiO2 ;
at 1500° (the temperature of firing porcelain), this sinters to a stony
mass, which softens at 1650°, and at 1700° fuses to a brown or grey
viscous liquid.

In order to separate from clay the oxide of iron which dis-
colours the product, Schwerin mixes clay with water and dips in
electrodes. The clay particles wander to the anode, the oxide of
iron to the cathode. The clay behaves in some ways like a colloid ;
brick clay becomes much more plastic if mixed with a little dilute
alkali, which appears to give charges to the clay particles, causing
them to repel one another. In ordinary brickmaking, the clay
is kneaded with water and allowed to stand, when organic colloids
(humic acids), conferring plasticity, appear to be formed. (An
infusion of straw has the same effect.) Clay used in making pottery
is washed, and the coarse particles are allowed to settle. The fine
clay is then allowed to deposit, and excess of water removed by air
drying. It is then highly plastic, and can be worked on the wheel.
The goods are air dried by stacking in warmed rooms, and then
burnt in clay boxes, called seggars, stacked in a kiln. ,The product,
which has undergone shrinkage, is called biscuit or earthenware. In
treating porcelain clay, the mass is sterilised before working up,
as further fermentation would develop bubbles. In the Berlin
porcelain works the sterilisation is effected by exposure to ultra-
violet light.

902 INORGANIC CHEMISTRY CHAP.

Bricks are made from impure clay, containing sand and oxide of
iron, which gives them a red colour after firing at about 950°.
The yellow bricks used in the South of England are made from mix-
tures of clay and chalk. Purer clay is used for earthenware, which is
fired at a higher temperature : 1-3 per cent, of Fe2O3 forms a buff-
coloured product ; 4-5 per cent. a. red. Porcelain is made from a
mixture of the purest China-clay, or kaolin, free from iron, with a
material containing silica. Thus, Berlin porcelain is made from
55 parts of kaolin, 22-5 of pure quartz, and 22-5 of felspar. It is
fired at about 900°, then the glaze is put on, and the goods are fired
at a bluish-white heat (1400-1500°). The temperature is regulated
by pyrometers, or by small clay cones (Seger cones), which soften
and bend over at particular temperatures in the furnace. The
mass undergoes partial fusion and the resulting product is trans-
lucent.

In the process of firing clay, the particles at the highest tem-
perature undergo partial fusion and become cemented together,
forming a stony mass. Clay containing a large proportion of silica
and alumina in comparison with the basic oxides (Na2O,CaO)
always present as impurities, is very refractory, and is called
fireclay (e.g., Stourbridge clay). This is made into refractory bricks,
and to prevent undue contraction on firing, broken firebricks
(" grog ") are added to the clay before heating. Graphite may also
be incorporated with the fireclay when it is formed into crucibles.

The clay after firing forms the body of the ceramic ; this is
called biscuit if porcelain clay is used ; otherwise it is called earthen-
ware. It is next glazed. The glaze is a glassy surface imparted to
the body, and intimately united with it. Earthenware drainpipes
and cheaper goods are often salt-glazed ; common salt is thrown into
the kiln and is vaporised at the high temperature, forming a thin
layer of fusible silicate on the surface of the ware. Salt-glazed ware
is suitable for pipes for conveying acids. Table-ware is usually
lead-glazed : the ware is dipped into a creamy paste of a mixture of
60 parts of lead oxide, 10 of clay, and 20 of ground flints. Some
of this adheres to the surface: and is fused in the furnace to a
glass. Porcelain is glazed by dipping and re-firing, as in the case of
earthenware. The glaze may be ground felspar, or mixtures ; e.g.,
Berlin glaze consists of : kaolin 31, quartz 43, gypsum 14, and
broken porcelain, 12.

The ware may be painted before glazing (some colours are applied
on the glaze) ; the colours are metallic oxides (e.g., cobalt oxide),
which form coloured glasses (p. 850) with the glaze, or with lead
oxide and silica, or borax, applied with the colouring oxide, before
the glaze is applied. In porcelain used in laboratories the glaze
must adhere firmly to the body, and the thermal expansions be so
adjusted that no tendency to separation occurs on heating. Berlin

XLIII METALS OF GROUP III OF THE PERIODIC SYSTEM 903

porcelain is well known for its excellence in these respects. English
bone, chin contains 30-50 per cent, of bone-ash (calcium phosphate).
It is less resistant than Berlin ware.

The following table contains a classification of ceramic products :

I. POROUS BODY, permeable to water :

(1) Unglazed (a) softens above 1400° (non- refractory) — terra cotta ;

(b) does not soften above 1400° (refractory) — firebrick,
refractory ware.

(2) Glazed (a) fine earthenware (white body) ;

(b) sanitary ware (fireclay body) ;

(d) Majolica (enamelled faience), first made in
Majorca.

II. NON-POROUS BODY, impermeable to water :

(a) translucent : porcelain ;

(b) opaque : stoneware.

Crucibles are made from a pure clay mixed with coarse sand or ground
burnt clay. The most refractory kinds contain the largest proportion
of silica. A mixture of clay and graphite is also used.

Ultramarine. — The rare mineral lapis lazuli, which has a beautiful
blue colour, is a sodium-aluminium silicate containing sulphur in
some form not completely denned, but probably as sodium sulphide.
Ancient Egyptian amulets of this stone (which is very soft) are
common. In 1828, Gmelin obtained artificial lapis lazuli, or ultra-
marine, by heating clay with sodium sulphate and carbon.

A mixture of 100 parts of kaolin, 70 of soda-ash, 80 of sulphur,
and 14 of resin is heated to bright redness in a closed crucible. A
white ultramarine, with the approximate composition Na7Al3Si3S2012,
is formed. If air is admitted during heating, a green ultramarine,
Na5Al3Si3S2012, is formed. If this, or white ultramarine, is mixed
with powdered sulphur, and heated in air. blue ultramarine,
Na4Al3SiaS2012, is formed, which is ground and washed. If this,
which is the commercial product, is heated in a stream of dry
chlorine, nitric oxide, or hydrogen chloride, a violet, and finally a
red, ultramarine result. The cause of the colours is not clear : it
has been suggested that colloidal sulphur is present.

Alkalies are without action on ultramarine, so that it can be
used in laundering to give a white appearance to linen, as it is not
attacked by soap or soda. Acids, however, rapidly decompose it,
with evolution of sulphuretted hydrogen and a white, gelatinous
residue remains. Fuming sulphuric acid does not produce this
change. The sodium in ultramarine may be replaced by its equiva-

904 INORGANIC CHEMISTRY CHAP.

lent of silver by treatment with silver nitrate, and a brown silver
ultramarine obtained. Potassium and lithium chlorides give, with
silver ultramarine, corresponding potassium and lithium ultra-
marines.

GALLIUM (Ga — 69-5), AND INDIUM (In == 113-9).

Gallium and Indium. — The rare element gallium (Ga = 69-5) occurs
in minute traces in most specimens of zinc blende, and was discovered
by the spectroscope in a blende from Pierrefitte by Lecoq de Boisbaudran
in 1875. It is the eka-aluminiwn of Mendel6eff (p. 470). Gallium
occurs in traces in bauxite, and in commercial aluminium. Middles-
brough cast-iron contains 1 part of gallium in 33,000. Gallium fuses
at 30-1° and remains supercooled, so that it is often considered as a
liquid element, along with mercury and bromine. Indium, In = 113-9,
was discovered by Reich and Richter in the spectroscopic examination
of zinc blende from Freiburg (1863). It gives a dark blue flame colora-
tion. The oxide is In2O3, but three chlorides, InCl, InCl2, and InCl3,
are known, with normal vapour densities. Indium and gallium form

III
alums, K2SO4,R2(SO4)3,24H2O

THALLIUM. Tl =-. 2024.

Thallium. — In 1861, Crookes observed a bright green line
in the spectrum of a specimen of flue dust from a vitriol works,
which he found was due to the presence of a new metal. The
element was independently discovered a year later by Lamy.
Crookes gave it the name thallium, from the Greek thallos, a young
twig, on account of the colour imparted to the flame. The only
minerals rich in thallium are crookesite (17 per cent. Tl, with Se, Cu,
Ag), and lorandite, TlAsS2.

Thallium may be obtained from vitriol flue-dust, or from pyrites
(from which it passes into the flue-dust), by dissolving in aqua regia,
evaporating, precipitating with sulphuretted hydrogen and then
ammonia in the usual group separations, and then adding potassium
iodide to the filtrate. A yellow precipitate of thallous iodide, Til,
is formed, which gives a green coloration when heated on platinum
wire in a Bunsen flame. If this is reduced with zinc and dilute
sulphuric acid the metal is obtained. Thallium is a soft, greyish-
white metal, m.-pt. 303° ; its vapour density corresponds with the
formula T12. It oxidises in moist air, decomposes steam at a. red-
heat, and dissolves readily in dilute sulphuric, and especially in
nitric, acid. It is less easily soluble in hydrochloric acid, since
thallous chloride, T1C1, is sparingly soluble.

Thallium forms two series of compounds : the thallous compounds,
RX, in which it is univalent and shows analogies with silver and the

XLIII METALS OF GROUP III OF THE PERIODIC SYSTEM 905

alkali-metals ; and the thallic compounds, RX3, in which it is ter-
valent, and exhibits resemblances to aluminium.

If thallium is dissolved in dilute sulphuric acid and the solution
evaporated, thallous sulphate, T12SO4, isomorphous with potassium
sulphate, and forming an alum, T12SO4,A]2(SO4)3,24H2O, is obtained.
From its solution, hydrochloric acid precipitates white thallous
chloride, T1C1, resembling silver chloride in becoming violet on
exposure to light, but differing from silver chloride in being sparingly
soluble in ammonia. With chloroplatinic acid a sparingly soluble
chloroplatinate, Tl2PtCl6, resembling K2PtCl6, is formed. * Iodides
precipitate yellow thallous iodide, Til, almost insoluble in cold
water, but dissolving in 800 parts of boiling water (cf. PbI2).

Thallous hydroxide, T10H,H20, is obtained in yellow needles
by decomposing a solution of thallous sulphate with baryta-water
and evaporating. The solution turns turmeric paper brown, and
is therefore alkaline (cf. KOH), but then bleaches it. If heated
out of contact with air at 100°, T1OH forms black thallous oxide,
T12O, dissolving in water to form a colourless solution of T1OH.
On addition of bromine and alkali, this solution gives a brown
precipitate of thallic hydroxide, T1(OH)3, or TIO(OH), which loses
water on heating and forms reddish-brown thallic oxide, T1203.
This evolves chlorine with hydrochloric acid, and forms T1C1
(cf. Pb203).

Sulphuretted hydrogen throws down a black precipitate of
thallous sulphide, T12S, from alkaline solutions of thallous salts.
It is soluble in dilute acids (except acetic), but insoluble in ammo-
nium sulphide.

Thallous hydroxide solution absorbs carbon dioxide, forming the
soluble thallous carbonate, T12C03, the solution of which is hydrolysed
(cf. K2C03).

Thallic chloride, T1C13.4H2O, is formed by passing chlorine through
thallous chloride suspended in water, and evaporating at 60°.
Thallic sulphide, T10S3, is a black pitch-like mass, obtained by fusing
thallium with excess of sulphur. Thallic sulphate, T12(SO4)3,7H2O,
is formed by dissolving thallic oxide in dilute sulphuric acid ; it is
decomposed by water with precipitation of a basic salt,
T1(OH)S04,2H20, and forms with potassium sulphate a compound
K2SO4,T12(SO4)3,8H2O, which is not a true alum.

In its analogies to the alkali -metals, lead, and aluminium, thallium
shows a greater diversity of properties than most other elements :
Dumas appropriately called it the " ornithorhynchus amongst the
metals " — the duckbill platypus.

Thallium is used to a limited extent in the production of a very
refractive optical glass, obtained by fusing the carbonate with sand and
red lead.

906 INORGANIC CHEMISTRY CHAP.

An oxide, T1O, is obtained as a black precipitate by the action of
hydrogen peroxide on an alkaline solution of thallous sulphate ; its
formula is considered to be T1-OT1:O. Another oxide, T13O5, is
said to be deposited on the anode in the electrolysis of a solution of
T12SO4 faintly acidified with oxalic acid.

THE RARE EARTHS.

The rare earths. — The substances known as the rare earths are
the oxides of metals which, with the exception of cerium, belong
to the third group of the Periodic System. Their general formula
is thus R20s ; the most stable cerium oxide, however, is Ce02.
They occur in rare minerals found in Scandinavia, Siberia, Green-
land, North America, and Brazil, usually in the form of silicates.
Not only are some of these elements present in small amounts
in the crust of the earth, but they differ from such rare elements
as lithium, which are widely diffused, in occurring solely in a few
special localities. Their compounds are therefore (with the excep-
tion of those of cerium) very expensive and were, until the fairly
recent discovery of the monazite deposits of Brazil and Carolina,
in the hands of a very limited number of chemists. The properties
of many members of this group of elements are consequently
imperfectly known. In addition to this, the different elements
resemble one another so closely, and are separated only with such
great difficulty, that many substances formerly thought to be
definite chemical individuals have on further investigation proved
to be mixtures, and in many cases the individuality of some of the
rare earths is still a matter of doubt. Crookes, to whom much of
the pioneering work on this group of elements is due, concluded
in 1887 that the elements contained in the rare earths might be
mixtures of closely related elements, the atomic weights of which
were very near together. He called these meta-elements, and
supposed that many of the ordinary chemical elements might be
of similar constitution. Improved methods of separation of the
rare earths have not confirmed Crookes's hypothesis, and the recent
work on the JT-ray spectra of the rare earths (p. 1030) has pfc
their individuality on a more satisfactory basis.

As an example of the difficulties encountered in this branch
chemistry, reference may be made to the separation of an earth calk
" didymia," regarded as a pure substance by Lecoq de Boisbaudrai
(1879), into two new earths, neodymia and praseodymia, by Welsbach in
1885. The " didymium " salts were colourless, but in solution exhibited
an absorption spectrum in the green and red. By repeated crystallisa-
tion of the nitrates from nitric acid, two fractions were obtained, one

XLIII METALS OF GROUP III OF THE PERIODIC SYSTEM 907

green (praseodymium salt) and the other rose-coloured (neodymium
salt), showing separately the two parts of the absorption spectrum of the
original substance. The colours are complementary, and the mixture,
as in the case of a mixture of cobalt and nickel salts, is colourless. Since
neodymia and praseodymia always occur with the other earths, the
absorption bands in the spectrum, even of light reflected from
the sand or native earth, is an indication of the presence of rare
earths.

The rare earths exhibit very beautiful phosphorescent effects
on exposure to cathode rays in vacuum tubes, and phosphorescence
spectra obtained in this way were studied by Crookes. It has been
found, however, that the pure earths are not phosphorescent, but
show the effect only in presence of small amounts of other sub-
stances, so that the importance once attached to these spectra has
receded.

Rare earth minerals. — Minerals containing the rare earths occur
in relatively few localities, and each mineral usually contains a
number of the earths. Cerite contains lanthanum, praseodymium,
neodymium. and samarium, in addition to cerium, and also traces
of other earths : gadolinite contains chiefly yttrium, erbium, etc.,
with only small amounts of cerium and lanthanum. The rare
earths are therefore usually divided into two groups :

I. Cerite earths : oxides of cerium, lanthanum, praseodymium,
neodymium, samarium, and europium.

II. Gadolinite earths : oxides of gadolinium, scandium, yttrium,
terbium, dysprosium, erbium, thulium, and lutecium.

The earths called celtia, phillipia, mosandria, decipia, and victoria
have been proved to be mixtures of the above. Examples of rare
earth minerals are the following : cerite, H3(Ca,Fe)Ce3Si3O13 ; orthite,
A10HCa2(Al,Fe,Ce)2(Si04)3 ; gadolinite, (Fe3e)2Y2Si2010 : xeno-
tine. YP04 ; fergusonite, YNbO4 ; Australian fergusonite, YTaO4 ;
columbite and tantalite, [(Nb,T'a)03]2(Fe,Mn) ; euxenite, polycras,
blomstrandite, and priorite, containing Nb. Ta, and Ti ; samarskite,
containing U, Th, Nb, Ta ; microlith,' Ca2(Ta5Nb)207 ; yttro-
tantalite, Y4(ta207)3.

Separation of the rare earths.— The rare earths are precipitated
by oxalic acid from acid solutions. The different earths so obtained
are then separated by one or more of the following processes : —

(1) Fractional decomposition of the nitrates by ignition.

(2) Fractional precipitation with a base.

(3) Fractional crystallisation of salts and double salts, e.g., with
ammonium nitrate, bismuth nitrate, etc.

(4) Fractional precipitation of salts with oxalic acid, succinic
acid, sodium stearate, etc. ,

A separation, e.g., by fractional crystallisation, may be represented

!)08 INORGANIC CHEMISTRY CHAP.

diagrammatical ly by Fig. 416, and from this the tedious character oi the

operation may be inferred. Many hundreds of fractions may be neces-

Crude Salt sary t° attain separation. In the diagram,

the less soluble constituent may be con-
sidered as accumulating on the right
hand side.

Cerium, Ce = 139-15. — The only rare
earth element of importance is cerium,
compounds of which are produced in
relatively large amounts in the pre-
paration of thorium salts from monazite
1 |15 (p. 930). By the electrolysis of the
no. 4i6.-Diagram illustrating chlorides of these elements, impure
separation of Bare Earths. metallic cerium (containing lanthanum

and other rare earth elements), which

is known as " Mischmetall," is obtained. This is alloyed with
iron, and used in automatic lighters, since when it is abraded
with steel it throws off very hot sparks which will ignite coal gas,
or the vapours of alcohol and petrol.

in
Cerium forms two series of compounds, viz., the cerous salts, CeX3.

and the eerie salts, CeX4. The cerous salts, in which the element
is tervalent, are stable and colourless, usually similar in com-
position and isornorphous with the corresponding compounds of
other rare earth elements. If, however, cerous salts with volatile
acids (oxalate, nitrate) are heated, the oxide remaining is not cerous
oxide, Ce203, corresponding with the rare earths, but cerium dioxide,
CeO2, which is the stable oxide and is known as ceria. Cerous oxide,
Ce203, is obtained by reduction of the dioxide with calcium. Cerous
hydroxide, Ce(OH)3, which is formed as a white precipitate on addi-
tion of alkalies to solutions of cerous salts, is rapidly oxidised on
exposure to air, becoming red and violet, and finally pure yellow
when eerie hydroxide, Ce(OH)4, is produced. The latter is obtained
by adding sodium hypochlorite and alkali to a solution of a cerous
salt.

Cerium dioxide, or ceria, Ce0.2, obtained by heating the oxalate,
is a nearly white powder with a faint yellow tinge. If traces of
praseodymium salts are present, the oxide is darker in colour, and
1 per cent, of Pr203 communicates to ceria a dark brown colour.
The commercial oxide is usually yellowish-brown. Ceria when
treated with hot concentrated sulphuric acid forms yellow eerie
sulphate, Ce(S04)2, which is a powerful oxidising agent and dissolves
in water to form a yellow solution. The solution is reduced, with
evolution of oxygen, by hydrogen peroxide, a colourless solution of
cerous sulphate, Ce2(SO4)3, being formed. This gives a charac-

XLIII METALS OF GROUP III OF THE PERIODIC SYSTEM 009

teristic double salt with potassium sulphate, Ce2(S04)3,K2SO4,3H2O,
which is insoluble in a solution of potassium sulphate, and is used
in the separation of cerium.- Cerous oxalate, Ce2(C2O4)3,10H.2O,
is precipitated from solutions of cerous salts by oxalic acid.

Ceria dissolves with difficulty in concentrated hydrochloric acid,
forming a dark brown unstable solution of eerie chloride, CeCl4,
which on heating evolves chlorine and leaves a solution of csrous
chloride, CeCl3. This is obtained anhydrous as a yellowish-white
sublimate when a mixture of ceria and carbon is heated in chlorine.

Ceric salts are hydrolysed by water, basic salts being precipitated.
The most stable are the double nitrates with alkali metals,
R2Ce(N03)6, which crystallise well, are soluble in water and alcohol,
and are bright red in colour.

If ceria is heated in hydrogen, a dark blue suboxide, Ce4O7, is formed,
which smoulders when heated in air, forming CeO2. A hydride, CeH3,
is formed as a dark blue powder when hydrogen is passed over cerium at
250-270°. It ignites spontaneously in air.

Cerous salts are oxidised to eerie salts by potassium perman-
ganate in neutral or slightly alkaline solution, by ammonium per-
sulphate in hot dilute solutions containing a little persulphuric
acid, or by anodic oxidation in electrolysis. Ceric salts are reduced
to cerous salts by electrolytic reduction, by prolonged boiling with
concentrated hydrochloric acid, or more rapidly by hydrochloric
acid and stannous chloride. Cerous salts in alkaline solution are
reducing agents : they precipitate cuprous oxide from Fehling's
solution : 2CuO -f- Ce203 = Cu2O + 2Ce02, and mercurous oxide
from mercuric chloride : 2HgO -f Ce2O3 = Hg2O -f 2CeO2. Gold
and silver are precipitated as metals. In this reducing action they
differ from the salts of all the other rare earths.

In alkaline solution, cerium salts are oxidised by hydrogen
peroxide to a reddish-brown hydrated peroxide :

Ce(OH)4 + H(O2H) = Ce(O2H) (OH)8 + H20.

If potassium carbonate and then H2O2 are added to a neutral solu-
tion of a cerous salt, and the liquid is warmed to 40-60°, a yellow
colour due to cerium peroxide is formed. This is a delicate test for
cerium.

Ce. La. Nd. Pr. Sa.

Specific gravity 7-0242 6-1545 6-9563 6-4745 7-7-7-8

Melting point 623° 810° 840° 940° 1300° — 1400°

Heat of combustion 1-603 1-602 1-506 1-467 kgm.cal. per gm.

Specific heat 0-04479 0-04485 — — —

910 INORGANIC CHEMISTRY CH. xuu

EXERCISES ON CHAPTER XLIII

1. How is alum prepared ? ' What chemical compounds are known
" alums " ?

2. In what forms does aluminium occur in Nature ? How is the
metal manufactured, and for what purposes is it used ?

3. For what purposes are aluminium salts used ? Describe the
preparation of (a) aluminium sulphate from kaolin, (b) anhydrous
aluminium chloride from aluminium. What is the molecular weight
of the chloride ?

4. How are aluminates obtained ? Describe reactions in which
alumina functions as an acidic and as a basic oxide.

5. How is porcelain made ? What varieties of ceramic products are
manufactured ?

6. Where do gallium, indium, and thallium occur ? How were these
elements discovered ?

7. How may a thallium salt be obtained from iron pyrites containing
this element ? What elements does thallium resemble ? Discuss its
position in the Periodic System.

8. What are the " rare earths " ? How are they separated from one
another, and for what purposes are any of them used ? Discuss the
position of cerium in the classification of the elements.

CHAPTER XLIV

THE METALS OF THE FOURTH GROUP

The carbon group — Group IV in the Periodic System, often
called the Carbon group, contains two non-metals, carbon and
silicon, and seven metals. The two sub-groups are :

Odd series : Even series :

Germanium, Ge = 71-9, m.-pt. Carbon, C = 11-91, m.-pt.

960°, sp. gr. 5-47. 7360°.

Tin, Sn = 117-8, m.-pt. 232°, Silicon, Si = 28-1, m.-pt.

sp. gr. 7-29. 1420°, sp. gr. 2-49.

Lead, Pb = 205-55, m.-pt. 327°, Titanium, T = 47-72, m.-pt.
sp. gr. 11-35. 1800-1850°, sp. gr. 4-87.

Zirconium, Zr = 89-9, m.-pt.

1530°, sp. gr. 4-08.
(Cerium, Ce = 139-15, m.-pt.

623°, sp. gr. 7-0).
Thorium, Th = 230-31, m.-pt.
1700°, sp. gr. 11-0.

Of these metals all but two, tin and lead, are rare. Cerium,
although forming compounds of the type CeX4, typical of the group,
is more conveniently described with the rare earths (Chapter
XLIIL). In Group IV, the differences between the odd and even
series are very ill-defined. The electrochemical characters of the
elements are also not pronounced, because the group forms the
transition between the electropositive (base-forming) elements of
group III, such as aluminium and the metals of the rare earths,
and the electronegative (acid-forming) elements of the succeeding
group V, such as nitrogen and phosphorus.

The two non-metals of the group are fusible only with the greatest
difficulty ; the metals also, with the exception of tin and lead, have
high melting points. Carbon, silicon, germanium, zirconium, and
thorium, form hydrides, RH4. All the elements of the group form

911

912 INORGANIC CHEMISTRY CHAP.

chlorides. RC14, although in the case of lead the stable chloride is
PbCl2 :

SiCl4, b.-pt. 57-5° ; SiHCl3, CC14, b.-pt. 76-7° ; CHC13,

b.-pt. 34°. b.-pt. 61-2°.

GeCl4, b.-pt. 86° ; GeHCl3, TiCl4, decomposes.

b.-pt. 75°. ZrCl4, sublimes.

SnCl4, b-pt. 114-1°. CeCl4, stable only in solution.

PbCl4, decomposes. ThCl4, m.-pt. 8-20°; sublimes.

Especially characteristic of the group are the compounds RHC13,
known as chloroforms.

The typical oxides, R02, are all known. Numerous other com-
pounds besides those corresponding with the type RX4 are formed
by the elements. In the cases of C, Si, Ge; Ti, Zr; Th, RX4 is the
stable type ; Sn and Pb form stable compounds of the type RX2 ;
in the case of lead, the only stable compounds of the quadrivalent
type are the dioxide, Pb02, and some double compounds. Cerium
also forms compounds of the type RX3, and on the whole shows
close analogies to elements of the preceding group. In many of its
chemical properties lead shows close analogies to barium, in the
second group ; e.g., its sulphate, PbS04j is very sparingly soluble in
water and is isomorphous with BaSO4, with which it often occurs
in the ores.

The element carbon differs from all the other elements in the
number of its compounds. The study of these constitutes a special
branch of chemistry — organic chemistry (p. 658).

TIN. Sn = 117-8.

Tin. — Although it is supposed that the word bedil in the Old
Testament refers to tin, the metal was first distinctly mentioned by
Pliny, who speaks of plumbum nigrum (lead), and plumbum can-
didum (tin), observing that the latter was brought from the Islands
of Cassiterides, in the Atlantic. This undoubtedly refers to the
British Isles, and the island Iktis, on the coast of Britain, which
(according to Diodorus Siculus) was separated from the mainland
only at high water, is no doubt St. Michael's Mount, Cornwall,
where tin ore is found. The metal was afterwards given the Latin
name stannum. The Latin Geber refers to the curious crackling
noise, or " cry of tin," resulting when a bar of tin is bent ; this is
due to the friction of the crystalline particles. The alchemists
associated tin with the planet Jupiter, giving it the symbol 11 : the
thunderbolt of Jove.

Tin occurs in small quantities in Siberia, Guiana, and Bolivia
in the metallic state ; its commonest ore is tinstone, or cassiterife,
the dioxide, Sn02 (m.-pt. 1127°), which is found in large quantities

THE METALS OF THE FOURTH GROUP

013

XL IV

in Devon and Cornwall, the Straits Settlement, Saxony, Peru, the
United States, Australasia, South Africa, and in other localities. It
occurs either massive or as an alluvial deposit (stream tin), and
crystallises in tetragonal prisms, terminated by pyramids (Fig. 417).
It is a dense mineral (sp. gr. 64-7-1), easily separated from lighter
rocks by washing. If necessary the ore is crushed, and washed in
a current of water, the process being known as huddling. If wolfram
(FeW04) occurs with the tinstone, it cannot be separated in this
way, since its density is 7-1— 7*9 ; recourse is then had to electro-
magnetic separation (p. 10). The total production of tin in 1913
was 120,300 tons.

Metallurgy of tin. — The ore, after " dressing," i.e., separation from
gangue, wolfram, etc., is first calcined in an inclined revolving
tube-furnace (Oxland and Hocking's calciner) (Fig. 324). The ore
is fed in at the top, and meets the flame and hot gas from a furnace
at the lower end. Sulphur and arsenic are
expelled as sulphur dioxide and arsenic tri-
oxide (As203), the latter being condensed in
flues. Copper and iron form oxides and sul-
phates. The calcined ore is discharged from
the lower end of the furnace ; it is cooled and
washed with water to remove copper sulphate,
which goes into solution, and ferric oxide and
light matter, which are washed away. The
treated ore, known as black tin, now contains
60-70 per cent, of Sn02. It is mixed with
one-fifth of its weight of ground anthracite
and a little fluor-spar, moistened, and
smelted in a reverberatory or a shaft furnace : Sn02 -}•- 2C =
Sn + 2CO.

The product is refined by liquation, i.e., by heating bars of the
metal on the hearth of a reverberatory furnace, when the readily
fusible tin (m.-pt. 232°) flows away, leaving a dross consisting of an
alloy of tin with copper, iron, and arsenic. The metal is finally
fused and " poled " with billets of green wood (p. 807), when the
remaining impurities separate as a scum. The scum and dross are
worked up by smelting. The tin is heated to 200°, when it becomes
brittle and can then be broken up by a hammer, yielding grain-tin.
On slowly cooling molten tin, crystals are formed. The crystalline
structure of the metal is destroyed on rolling, tinfoil being produced.

Properties. — Metallic tin has a bright white colour, and, after
fusion, a specific gravity of 7-30. The metal is very fusible, but
has a high boiling point (2270°). Its lustre is not impaired by ex-
posure to air or water, either separately or conjointly, whereas lead
is attacked. For this reason tin is used for tinning copper or iron
vessels. These are first of all thoroughly cleaned, heated, and then

3 N

T i

1 j

! 1

1 1

i i

4~~-JL

/ —

Fio. 417.— Crystal of
Tinstone.

914 INORGANIC CHEMISTRY

molten tin is poured in. This is brushed over the surface of the
other metal, rosin and salammoniac being added as fluxes. Tinplate
is made by dipping clean sheets of iron (given a bright surface by
" pickling " in sulphuric acid) into molten tin. covered with melted
palm oil. The sheet then passes under a partition in molten
tin covered with melted fat, and then through rollers to remove
superfluous metal.

Tin is recovered from scrap tinplate by the detinning process.
The material is washed with alkali to remove grease, rinsed and
dried, and heated to melt off the solder. The metal is then treated
with chlorine gas in iron cylinders, kept cool. Volatile stannic
chloride. SnCl4, is formed, and the residue of iron scrap, containing
less than 0-1 per cent, of tin, is hydraulically pressed into blocks and
smelted in the open-hearth furnace (p. 981).

When ordinary white tin is strongly cooled, it crumbles down to a
grey powder, of density 5-8. The transformation, is quickest at
— 50°. Grey tin is an enantiotropic form, the transition point,
Sna ^± Sn/?, being 18°. White tin is thus a metastable form under
ordinary conditions ; transformation occurs in contact with a
little grey tin, or a solution of stannous chloride. Granulated tin,
added to the latter, falls to a grey powder. White tin exists in two
allotropic forms. From 18° to 170° ordinary tin is stable, and
crystallises in the tetragonal system. At 170° transition into a
rhombic form, sp. gr. 6-5, occurs :

18° 170°

Grey tin ^± Tetragonal tin ^ Rhombic tin
sp.gr. 5-80 sp.gr. 7-286 sp. gr. 6-56

Tin oxidises when fused in the air, a grey scum or dross forming
on the surface. This consists of a mixture of tin dioxide and
unchanged tin : on heating in air it is converted into tin dioxide,
Sn02, which is yellow when hot, but becomes white on cooling
(" putty powder "). At a white heat tin burns in air with a white
flame.

Tin is only slowly attacked by dilute acids, but readily dissolves
in hot concentrated hydrochloric acid, forming a solution of stannous
chloride: Sn + 2HC1 = Sn012 + H2.

Dilute sulphuric acid slowly forms stannous sulphate, SnS04,
with evolution of hydrogen : hot concentrated sulphuric acid gives
the same salt and S02.

Concentrated nitric acid, when perfectly free from water, has
no action, but in presence of a trace of water it acts vio-
lently on tin, producing red fumes, and forming a small quantity
of soluble tin salt, and an abundant white residue of metastannic
acid, H2Sn5On (?). Boyle (1670) remarked that " aqua fortis
eats up more tin than it "dissolves." Hot alkalies dissolve tin with
evolution of hydrogen, forming stannates.

XLIV THE METALS OF THE FOURTH GROUP 915

Tin forms important alloys, e.g., bronze (p. 810). A mixture of

1 part of tin and 2 parts of lead is ordinary soft-solder (fine-solder
consists of equal parts of tin and lead). Pewter contains 4 parts of
tin and 1 part of lead, usually with a little antimony. Britannia
metal, a white metal, consists of 84 tin, 10 antimony, 4 copper, and

2 bismuth. Mirrors are sheets of very clean glass backed by pressing
them on a surface of amalgamated tinfoil. Tin forms with copper
the definite compounds Cu3Sn and Cu4Sn. Phosphor tin is a white,
metallic, coarsely crystalline mass, formed by adding phosphorus to
molten tin ; it melts at 370°. The definite compound SnP is known.
By adding phosphor tin to molten copper, phosphor-bronze is pro-
duced (p. 810).

Tin forms two series of compounds : the stannous compounds,

II IV

SnX2, and the stannic compounds, SnX4. These correspond with
the oxides, SnO and SnO2. The stannous compounds readily pass,
by oxidation, into compounds of quadrivalent tin.

Stannous compounds are therefore reducing agents. A solution of
stannous chloride when added to a solution of mercuric chloride gives
first a white precipitate of calomel, and if added in excess a grey pre-
cipitate of metallic mercury :

SnCl, + 2HgCl2 = SnCl4 -f 2HgCl
SnClJ + 2HgCl =: SnCl4 + 2Hg.

If stannous chloride is added to a solution of ferric chloride and
potassium ferricyanide, an immediate precipitation of Prussian blue
occurs, owing to the reduction of the ferric salt to a ferrous salt :
2Fev> + Sn" = 2Fe" -f SiT"

The lower oxide, stannous oxide, SnO, is basic, but the dioxide,
SnO2, shows feebly acidic properties, forming salts called stannates,
which are largely hydrolysed in solution : NaJSn03 -J- 2H20 ^±
2NaOH -f- H2SnO3- (stannic acid). In solution, the stannous salts
ionise, with formation of Sn" ; stannic salts usually form complex
ions, so that the existence of Sn"" is doubtful.

Stannous compounds. — Tin (e.g., tinfoil or granulated tin) readily
dissolves in hot concentrated hydrochloric acid, a solution of
stannous chloride being produced : Sn + 2HC1 = SnCl2 + H2. On
evaporating and cooling, the solution deposits transparent mono-
clinic prisms of SnCl2.2H2O, which melt at ,40°. They lose acid
on heating, and the anhydrous salt is best prepared by passing
hydrogen chloride over heated tin. It is soluble in alcohol and
ether, melts at 250°, and boils at 606°, the vapour being associated :
Sn2Cl4 — 2SnCl2. In solution in urethane, the substance has the
formula SnCl2. The crystals of hydrated chloride, known as tin
salt, do not give a clear solution unless hydrochloric acid is added ;
with water alone white stannous oxychioride, Sn(OH)Cl, is deposited.

3x2

<)!<'. INORGANIC CHEMISTRY CHAP.

Unless granulated tin is added, the acid solution quickly becomes
turbid from oxidation, stannous oxychloride being deposited, and
stannic chloride remaining in solution : GSnCl2 + 2H2O + O2 =
2SnCl4 + 4Sn(OH)Cl. With concentrated hydrochloric acid, crys-
talline hydrochlorostannous acids, HSnCl3 and H2SnCl4, are formed.
These form stable crystalline salts, e.g., (NH4)2SnCl4. Several com-
pounds of SnCl2 with ammonia are known.

If a piece of zinc is suspended in a solution of stannous chloride,
a bright crystalline deposit of tin is formed ("tin tree "). Large
crystals of tin are produced by adding zinc dust suspended in water
to a solution of stannous chloride.

Stannous bromide, SnBr2, is a light yellow salt, similar to the chloride.
Stannous iodide, SnI2, is a red crystalline substance, sparingly soluble
in water, but dissolving in hydriodic acid, or iodides, to form hydriodo-
stannous acid, HSnI3, or its salts, respectively.

Stannous sulphide, SnS, is formed as a brown precipitate when
hydrogen sulphide is passed through an acidified solution of stannous
chloride, or as a grey crystalline mass on heating tin with sulphur.
The brown precipitate is soluble in hot concentrated hydrochloric
acid (arsenic trisulphide is insoluble, cf. p. 655) ; it is not dissolved
by alkali-sulphides if these are perfectly free from excess of sulphur,
but dissolves readily in the polysulphides, e.g., yellow ammonium
sulphide. -It then forms first of all yellow stannic sulphide, SnS2,
which dissolves in the sulphide to produce a thiostannate, e.g.,
(NH4)2SnS3, from which acids re-precipitate stannic sulphide :
(NH4)2SnS3 + 2HC1 = 2NH4C1 + H2S + SnS2. The salt
Na2SnS3,2H20 is formed by boiling tin and sulphur with caustic
soda solution.

Tin dissolves slowly in dilute sulphuric acid, forming stannous sulphate,
SnSO4 ; a mixture of 1 vol. of H2SO4, 2 vols. of HNO3, and 3 vols. of
water may be used as a solvent. It dissolves in nitric acid diluted with
1 J- 2 vols. of water, forming stannous nitrate and ammonium nitrate :
4Sn -f 10HNO3 = 4Sn(NO3)2 + NH4NO3 + 3H2O. On strong cool-
ing, the solution deposits Sn(NO3)2,20H2O.

If caustic potash is added to a solution of stannous chloride, a
white precipitate of hydrated stannous oxide, 2SnO,H20, is produced.
On heating at 80°, this loses water and forms stannous oxide, an
olive-green powder also formed by heating stannous oxalate. This
smoulders when heated in air. forming the dioxide, Sn02. The
precipitate of hydrated oxide is soluble in excess of alkali, forming a
solution which appears to contain a stannite, H-SnO-ONa, analogous
to sodium formate : CO + NaOH = H-GO-ONa. The solution
has strong reducing properties ; e.g., it reduces a solution of cupric
sulphate to copper suboxide, Cu40.

XLIV THE METALS OF THE FOURTH GROUP 017

Stannic compounds. — When tin is treated with excess of chlorine
gas in a retort a volatile, strongly fuming, colourless liquid is pro-
duced. This is stannic chloride, SnCl4, which was discovered by
Libavius in 1605, and was called spiritus fumans Libavii. He
obtained it by distilling tin with corrosive sublimate : Sn + 2HgCl2
= 2Hg -f SnCl4. The vapour- density of stannic chloride (b.-pt.
114-1 °) corresponds with the formula SnCl4. With a small quantity
of water it dissolves with evolution of heat, forming a clear solution
from which the crystalline hydrates, SnCl4,3H2O, SnCl4,5H2O and
SnCl4,8H20, are obtained. The liquid also contains unchanged
SnCl4, which is volatile in steam. The hydrate SnCl4,5H2O is
prepared in commerce, and is called " oxymuriate of tin," or " butter
of tin." Stannic chloride is obtained in detinning scrap tinplate
(p. 914). The hydrate is used as a mordant, especially for silk, and
in " weighting " the latter. By treating SnCl4,5H2O with hydro-
chloric acid gas at 28°, and cooling to 0°, crystals of hydrochloro-
stannic acid, H2SnCle,6H2O, are formed, melting at 28°. Direct
combination of stannic chloride with alkali-chlorides gives chloro-
stannates, e.g., (NH4)2SnCl6, which crystallises anhydrous, and was
formerly used as a mordant in dyeing madder-reds and pinks
(hence it was called " pink salt "), until superseded by SnCl4,5H20.
The compound SnCl4,4NH3 is formed directly ; it can be sublimed
and dissolved in water without decomposition. The compounds
SnC)4)2SCl4? SnCl4,N203, SnCl4,2NOCl, SnCl4,PCl5, and SnCl4,POCl3
all readily formed directty.

Stannic bromide and iodide are formed directly. The fluoride, SnF4, is
formed from SnCl4 and anhydrous HF.

SnCl4. SnBr4. SnI4. SnF4.
M.-pt. — 33° 33° 146° sublimes
B.-pt, 114-]° 201° 295° 705°
Density 2-234(15°) 3-349(35°) 4-696 4-78
Colourless, White, Red, stable, White,
strongly fuming, octahedral deliquescent
fuming crystalline crystals. crystals-
liquid, solid

The fluoride forms complex salts, e.g., K2SnF6, analogous to silico-
fluorides.

Solutions of halogen compounds of quadrivalent tin contain the
un-ionised substances and their hydrolysis products, e.g., colloidal
stannic hydroxide, Sn(OH)4 ; the solution in hydrochloric acid
contains the ion SnCl6" : and it is doubtful if the stannic ion, Sn"",
is ever present as such, although Sn(OH)4 dissolves in sulphuric acid,
hydroxyl probably being eliminated in stages ; Sn(OH)4 -> Sn(OH)3'
->Sn(OH)2"-> Sn(OH) '"-> Sn"". From dilute solutions of stannic

918 INORGANIC CHEMISTRY CHAP.

salts the hydroxide separates as a gelatinous precipitate, especially
on boiling : SnCl4 + 4H2O ^± Sn(OH)4 + 4HC1. If the gelatinous
form is digested with a solution of potassium sulphate, it becomes
granular, filters readily, can be washed, and on ignition forms the
dioxide, Sn02.

Stannous compounds are oxidised and can be estimated by titration
with standard iodine t SnCl2 + I2 + 2HC1 = SnCl4 + 2HI, or ferric
chloride : Sn" -f 2Fe"" = Sn"" -f 2Fe'-: (p. 255). Stannic compounds
are usually estimated by precipitation of the sulphide, SnS2, which is
ignited, and the stannic oxide weighed.

Stannic acids. — The existence of at least two varieties of stannic
acid was the first case of isomerism recorded (Berzelius, 1817).
Colloidal stannic acid, formed in solutions of stannic chloride in
water, readily gelatinises. The precipitate is soluble in excess of
caustic potash or soda, a solution of a stannate, largely hydrolysed
and therefore alkaline, being formed : Sn(OH)4 + 2NaOH ^±
Na2SnO3 + 3H2O. From the solution, by evaporation, crystals of
sodium stannate, Na2Sn03,3H20, are obtained. Acids throw down
from this gelatinous a-stannic acid, which on drying at 100° has
the composition H2Sn03, and is soluble in dilute acids or alkalies.
The solution in dilute hydrochloric acid is identical with a solution
of stannic chloride in water. On standing, this solution slowly
deposits /3-stannic -acid (q.v.), which is probably a polymer of the
a-acid.

Sodium stannate, Na2Sn03,3H2O, used as a mordant, is prepared
by fusing tin dioxide with caustic soda, extracting with hot water,
and crystallising. The ignited dioxide, or the mineral tin-stone, is
insoluble in all acids except hydrofluoric, and does not dissolve in
aqueous alkalies. It can be brought into solution only by fusion
with caustic alkalies or their carbonates.

If tin is treated with fairly concentrated nitric acid, stannous
nitrate. Sn(N03)2, appears first to be formed. This is at once
oxidised by the nitric acid to stannic nitrate, Sn(N03)4, which can be
quickly separated if 70 per cent, acid is employed, but usually
undergoes hydrolysis. The final product is a white, curdy powder,
which is a stannic hydroxide, but differs from a-stannic acid in
being insoluble in dilute acids. It is slightly soluble in water, and
the solution reddens litmus. This variety of stannic hydroxide is
called /8-stannic acid, or metastannic acid. It was formerly sup-
posed to have the formula H2Sn5O11, but the proportion of water is
variable, and the difference between the a- and /?-acids seems to be
due to something more than varying hydra tion. If /3-stannic acid
is warmed with concentrated hydrochloric acid, it is converted into a
gelatinous solid hydrochloride, which, on pouring off the hydrochloric
acid and adding water, dissolves. /8-stannic acid is quickly deposited

XLIV THE METALS OF THE FOURTH GROUP 919

from the solution, especially on boiling, or on adding sulphuric
acid. Cold solutions of alkalies dissolve /3-stannic acid, forming
metastannates (e.g., Na2Sn5On,4H20, a sparingly soluble crystalline
powder), from solutions of which acids reprecipitate /?-stannic acid.
But if /3-stannic acid is fussed with alkali, an a-stannate, from which
acids throw down a-stannic acid, is produced.

Colloidal a-stannic acid is formed by dialysing a mixture of stannic
chloride solution and potash, or sodium stannate and hydrochloric
acid. As the electrolytes pass out, the gelatinous mass first produced
gradually forms a clear solution in the dialyser. On heating, /3-stannic
acid is precipitated.

If /3-stanm'c acid is treated with concentrated hydrochloric acid, a
gelatinous mass is produced, which is partly soluble in water. Hydro-
chloric acid added to the filtrate throws down a white precipitate, which
on drying in a vacuum has the composition Sn5O9Cl2,4H2O. It is a
glassy mass, soluble in dilute hydrochloric acid, but reprecipitated
by the strong acid. It is called /3-stannyl chloride, but may be a
salt of 0-stannic acid, which behaves as a weak base. The white
powder obtained by the action of concentrated nitric acid on tin may
be the corresponding nitrate.

If /3-stannic acid is heated with water at 100°, it passes into
another form, called para-stannic acid, H2Sn5On,2H2O (instead of
H2Sn5On,4H2O, which is supposed to be /3-stannic acid). The identity
of these compounds is, however, very ill-defined. Metastannic acid
absorbs phosphoric acid almost quantitatively from solutions, and may
be used in the separation of this acid in qualitative analysis.

Ferstannic acid corresponds with the unknown peroxide, SnO3. By
grinding SnO2 with H2O2, and drying the residue at 70°, the compound
HSnO4,2H2O is obtained ; if dried at 100°, H2Sn2O7,3H2O is formed.
By treating a stannate in the same way, perstannates, e.g., KSnO4,2H2O,
are formed.

Stannic sulphide, SnS2. — This compound is formed as previously
described (p. 916), or by precipitating a solution of a stannic salt
with H2S. It unites with alkali -sulphides to form thiostannates
(loc. cit.). The precipitate with H2S is yellow, but becomes black on
drying ; it is a mixture of the dioxide and disulphide. By adding
an acid to a solution of a thiostannate, free thiostannic acid, H2SnS3,
is precipitated, which on heating is converted into golden-yellow
SnS2. Crystalline SnS2 is obtained as a residue of golden-yellow glist-
ening scales (sp. gr. 4'425) (mosaic gold) by heating a mixture of tin
filings, sulphur, and salammoniac. It is. in this form, insoluble in
acids or alkalies, but dissolves in aqua regia :

Sn + 4NH4C1 = (NH4)2 SnCl4 + H0 + 2NH3
2(NH4).,SnCl4 -f S2 - SnS2 4- (NH4)2 SnCl6 + 2NH4C1.

920 INORGANIC CHEMISTRY CHAP.

LEAD. SEVERAL VARIETIES (p. 462). ORDINARY, Pb^- 205-55.

Lead. — The metal lead, which is easily reduced from its ores, is
mentioned in Job xix ; it was at first confused with tin, but the
difference was recognised by Pliny (cf. p. 912). The dull, heavy,
metal was associated by the alchemists Avith the slow-moving
planet Saturn, and designated by the symbol of the scythe, h.

Lead is widely distributed in the mineral kingdom ; traces occur
in the native form, but the chief ore is galena, the sulphide PbS,
which is a heavy (sp. gr. 7-5) mineral with a bright lustre, found in
many parts of the United Kingdom, especially in the north midlands
(e.g., Derbyshire) and south-western (Devonshire) counties ; it also
occurs in Flintshire, and at Leadhills in Scotland. Galena is
found in almost every part of the world. It is generally associated
with quartz, calcite, fluorite, and barytes, and usually contains
0-01-0-1 per cent, of silver. The oxides PbO and Pb304 (plattnerite)
are rare minerals ; and the carbonate cerussite (PbC03) : chloro-
phosphate pyromorphite (3Pb3,(PO4)2,PbClo) ; sulphate, anglesite
(PbSO4); sulpho-carbonate (leadhillite) (3PbC03,PbS04) ; and basic
sulphate, lanarkite (PbO,PbS04) occur less abundantly than galena.

Metallurgy of lead. — Lead is produced from galena by simple
roasting in an oxidising atmosphere ; its extraction was carried on
in England during the Roman occupation, and smelting in Derby-
shire was in active operation in the eighteenth century ; these
mines which were long abandoned, are at present being worked.
The process is carried out largely in reverberatory furnaces known as
Flintshire furnaces, introduced in 1698. The ore is first roasted at
a moderate temperature, when a portion of the galena is oxidised to
oxide and sulphate :

(1) 2PbS 4- 302 = 2PbO + 2SOa.

(2) PbS + 202 =-- PbSO4.

The temperature is then raised, a little quicklime added, and the
smelting reaction takes place, the remaining lead sulphide reacting
with the two oxidised products :

(3) PbS + 2PbO = 3Pb -f SO2.

(4) PbS + PbS04 ^r 2Pb + 2SO2 (reversible).

With the exception of about 10 per cent., which passes into the slag,
all tfie lead is obtained in the form of metal. The slag is afterwards
worked up by heating with lime and powdered coal, either in a small
blast furnace, or on the now nearly obsolete Scotch hearth, a flat
hearth with a tuyere for providing the blast.

Poorer ores, and an increasing amount of richer ores containing
quartz, blende, and pyrites, are now smelted in small blast furnaces.
The ore is first roasted (together with lime), and mixed with coke, old
slag, and a flux (consisting of iron pyrites containing silver and gold,

xi.iv THE METALS OF THE FOURTH GKOUI' 921

which pass into the lead). The lead oxide is reduced by the coke
and carbon monoxide, the sulphide by the iron : PbS -f- Fe =•
FeS + Pb, the sulphate by the sulphide and carbon, and the silicate
by carbon and lime or ferrous oxide.

Lead fume (chiefly PbO), formed during smelting, is collected in
flues and bag-filters, or by the electrostatic precipitation process.

The crude lead contains copper, antimony, and bismuth, which
render it hard . It is softened by melting on the hearth of a reverber-
atory furnace until the foreign metals are oxidised, and form a scum
on the surface, mixed with a little litharge (PbO). Lead is also
refined by electrolysis (Setts' process) in a solution of lead silico-
fiuoride. with a little gelatin, when a coherent deposit is formed.

Properties of lead. — Lead, if perfectly pure, has a silver-white
lustre, but has usually a bluish-grey colour. It is very soft, dense
(sp. gr. 11-35), and fusible (m.-pt. 327°). It is plastic, especially
when heated, when it may be Ci squirted " into wire by forcing it
through a die under pressure, or " wiped " in forming pipe- joints in
plumbing. Tubing is also formed by squirting. The so-called
" compo " tubing contains tin. Octahedral crystals of lead are
obtained by fusing the metal and allowing to cool, or by precipitat-
ing it from a solution of the acetate or nitrate by zinc (" lead tree ").
Monoclinic crystals a.re formed by electrolysis. Colloidal lead is
produced by reducing a solution of the chloride with hydrazine in
the cold. The metal boils at 1140° in a nearly perfect vacuum,
the vapour is monatomic at 1870°.

Lead oxidises rapidly but superficially in moist air, a white film
of hydroxide and carbonate being deposited. Pyrophoric lead,
obtained by heating the tartrate, ignites spontaneously in air (p. 166).
The metal is not attacked by pure water (except at the boiling point),
or by dry air, but is rapidly corroded by water containing dissolved
air ; the first product appears to be hydrated plumbous oxide,
Pb2O,2H2O, which rapidly oxidises, forming a loose deposit of
plumbic hydroxide. Pb(OH)2, which is appreciably soluble in water,
rendering the latter poisonous.

During the action of water containing dissolved oxygen 011 lead,
hydrogen peroxide is produced : Pb + 2H2O -f- O2 = Pb(OH)2 -f H2O2
— an example of autoxidation (p. 342). (On the solvent action of water
on lead, seep. 211.)

The atomic weight of ordinary lead determined by conversion of
the chloride into silver chloride, is 205-55 (H = 1) -/more than one
variety of lead, however, exists (cf. pp. 462 and 1033).

Lead dissolves readily in dilute nitric acid, or in hot concentrated
sulphuric acid, forming salts of the bivalent ion, Pb",, which is
colourless, and resembles the barium ion, Ba". in many ways. It is a
powerful cumulative poison, i.e., small quantities below the poisonous

922 INORGANIC CHEMISTRY CHAP.

snt

dose accumulate in the system, and ultimately induce chronic
poisoning. A characteristic symptom of lead poisoning, to whi
painters, plumbers, and potters using lead glazes are subject, is
blue line on the edges of the gums.

Several complex ions are known, and lead may form a constituen
of anions. Thus, if an alkali is added to a solution of lead salt, a
white precipitate of lead hydroxide is formed. This readily dissolves
in excess of alkali, forming a solution of a plumbite, K2PbO2 or
KHPbO2, which gives the anions Pb02" and HPbO2'5 but is largely
hydrolysed and reacts alkaline : Pb02" -f H2O ^ PbO + 20tf.
Ammonia does not dissolve lead hydroxide.

Lead oxides. — The following oxides oi lead are known :

(1) Pb2O — lead suboxide, formed as a black powder by heating the
precipitated oxalate below 300° : 2PbC2O4 = Pb2O + CO + 3C(32. It
is decomposed by heat, acids, or alkalies into Pb and PbO. Lead also
dissolves in a solution of the acetate, forming a sub-salt : Pb" + Pb =
2Pb\

(2) PbO — lead monoxide, basic, ordinary litharge, or massicot.

(3) Pb2O3 — lead sesquioxide. obtained by adding sodium hypochlorite
to a cold solution of PbO in caustic potash. It is a reddish -yellow
amorphous powder, decomposed by dilute acids into PbO (soluble) and
PbO2 (insoluble), hence it is probably a metaplumbate of lead :
PbO,Pb02

(4) Pb304 — red-lead or minium, also decomposed by acids, and pro-
bably 2PbO,PbO2, lead orthoplumbate.

(5) PbO2 — lead dioxide ; weakly basic and acidic, forming salts,
PbX4, and plumbates, e.g., K2PbO<, and Ca2PbO4.

Lead monoxide, PbO. — This oxide is formed on heating lead in air.
The grey dross so produced, which consists of a mixture of lead
monoxide and metallic lead, if heated in an iron vessel, turns yellow,
forming the monoxide PbO. The resulting yellow powder (which
darkens on heating) is called massicot ; if fused, and powdered,
the reddish -yellow crystalline form known as litharge is obtained.
Litharge, obtained in the refining of silver (p. 819), is largely used in
making flint-glass (p. 850), in glazing pottery (p. 902), in preparing
lead salts, and in making paints and varnishes. It accelerates
catalytically the absorption of oxygen by linseed oil, causing the
latter to " dry," or form a solid oxidised compound called linoxyn.
If litharge is . boiled with water and olive-oil, lead oleate, which is a
sticky adhesive mass used in making lead-plaster, is formed, and
glycerin passes into solution. Olive oil is triolein. or an oleic
ester of glycerin, and is saponified by the lead hydroxide,
traces of which are formed by the solution of the litharge (c/. p. 776).

Lead hydroxide, which appears to have the formula 2PbO,H20
or PbO(02H)s, is formed as a white precipitate on adding an alkali

XLIV THE METALS OF THE FOURTH GROUP 923

to a solution of a lead salt. It is slightly soluble in water
(as is PbO, which first forms the hydroxide), and the solution
turns red litmus blue. It appears to ionise as : Pb(OH)2 ^±
Pb(02H) " -f- OH'. It dissolves both in acids and bases, forming lead
salts and plumbites, respectively. The hydroxide loses water at
145°, forming the monoxide.

Red-lead, or minium, Pb304. — This important compound is
formed by roasting litharge in air at about 400°, and forms a scarlet
crystalline powder. It decomposes at 470° : 2Pb3O4 ^± 6PbO -f O2.
Red lead is used as a pigment, and in making flint glass.

Lead dioxide, Pb02. — When red lead is treated with concentrated
nitric acid, it is decomposed into lead nitrate and lead dioxide (or
lead peroxide) : Pb3O4 -f 4HNO3 --- 2Pb(NO3)2 + Pb02 + 2H2O.
On washing out the nitrate with water, chocolate-brown lead
dioxide, Pb02, remains. This oxide is always produced when lead
compounds are subjected to the action of powerful oxidising agents
in presence of alkalies. Thus, it is formed in an impure state when
bleaching-powder, or sodium hypochlorite, is added to lead monoxide
in alkaline solution : PbO + NaOCl ^ PbO2 + NaCl. Lead dioxide
is also deposited on the anode when an acid solution of a lead salt
is electrolysed between platinum electrodes ; pure PbO2 is formed,
and in this way lead may be separated from metals such as copper,
which deposit as such on the cathode. A lead plate used as an
anode in dilute sulphuric acid is oxidised by the SO/ ion discharged :
PbSO4 -f 2H2O + SO4 = PbO2 -f 2H2S04. To bring about the
discharge of the S04" ion, 2 x 96.000 cmb. must pass round the
circuit from anode to cathode : SO4" -f 2 x 96,000 cmb. = S04.
If this quantity of electricity passes round the circuit in the opposite
direction (from cathode to anode), the reaction is reversed : Pb02 +
2H2SO4 - PbSO4 + 2H2O + SO4 + 2 x 96,000 cmb. This is
the principle of the lead accumulator. A lead plate, pasted with a
mixture of red-lead and concentrated sulphuric acid, and another
plate of lead, are immersed in dilute sulphuric acid. On passing a
current from a dynamo, the lead sulphate at the anode is converted
into Pb02 by the above reaction. This is the operation of charging
the cell, and involves an expenditure of energy, in order to oppose
the E.M.F. of polarisation, about 2 volts. On discharging, both
plates are covered with lead sulphate. On recharging, the sulphate is
converted into the dioxide on the anode, and spongy lead on
the cathode.

The reactions in the accumulator are : —

+ anode : PbSO4 + 2H2O + SO/ - 2 x 96,000 cmb. ^ Pb02
+ H2S04.

- cathode : PbS04 ;=± Pb -f SO/ -f 2 x 96,000 cmb.

The upper arrows denote the charging, the lower arrows the dis-
charging, reactions.

924 INORGANIC CHEMISTRY CHAP.

It is not electricity or electrical energy which is stored in the cell,
but chemical energy ; the material PbSO4 is converted by the expen-
diture of electrical energy on the cell (leading to the chemical reactioi
of charging) into the two materials of a primary cell. In the Daniel
cell the energy was spent outside the cell in the reduction of the zii
ore to metallic zinc in the smelting process. But whereas the reactk
in the Daniell cell is not conveniently reversed by an electric current,
so as to put the cell into its initial active form, that in the accumulator
is easily reversed by an electrolytic method, with an expenditure of
energy practically the same as that obtained in the action of the cell.
The latter, therefore, acts as a reservoir of energy.

Lead dioxide is a powerful oxidising agent. A mixture of the
dioxide and sulphur ignites on trituration, burning with a brilliant
flame, and forming fumes of lead sulphate. Lead dioxide becomes
red hot when exposed to sulphur dioxide, and lead sulphate is pro-
duced : Pb02 -f- S02 = PbS04. If a manganous salt (e.g., MnS04)
is boiled with concentrated nitric acid, lead dioxide, and a little
dilute sulphuric acid, a pink solution of permanganic acid is formed.
This is Volhard's test for manganese : 2MnSO4 -j- 5Pb02 -f 3H2S04
= 2HMnO4 + 5PbS04 -f 2H2O. Chromic hydroxide, in presence
of an alkali, is oxidised to a chromate.

Plumbates. — If litharge and quicklime are heated together in air,
the mass takes up oxygen, forming calcium plumbate, Ca2Pb04,
or 2CaO,I*b02: 4CaO + 2PbO + O2 = 2Ca2Pb04. This may be
obtained in nearly colourless crystals, Ca2PbO4,4H20. Calcium
plumbate was the intermediate product in Kassner's oxygen process
(p. 169). A similar reaction occurs on adding lead dioxide to 100
gm. of caustic potash and 30 gm. of water fused in a silver dish ;
from the solution in water containing excess of alkali, crystals of
potassium plumbate, K2PbO3,3H2O, or K2Pb(OH)6, are deposited by
evaporating in a vacuum and adding a crystal of the isomorphous
stannate. These two salts are derived from orthoplumbic acid, H4Pb04
or Pb(OH)4, and metaplumbic acid, H2PbO3, respectively. The
former is not known in the pure state ; the latter is deposited as a
black powder on the anode on electrolysing a slightly alkaline solu-
tion of sodium lead tartrate.

II IV

Minium, or red lead, may be regarded as lead orthoplumbate, Pb2PbO4;

II IV

lead sesquioxide as lead metaplumbate, PbPbO3 — the sesquioxide is in
fact formed on precipitating a lead salt with a solution of a plumbate.
When calcium plumbate is heated to 250° in dry air, a perplumbate,
CaPb2O6, is said to be formed.

Halogen compounds of lead. — Two series of halogen compounds,

II IV

PbX2, and PbX4, are known. These may be called plumbous, and

XLIV THE METALS^QE THE FOURTH GROUP 925

plumbic, compounds, respectively, although the true plumbous
compounds correspond with Pb2O. Lead dichloride, plumbous
chloride, or simply " lead chloride," PbCl2, is slowly formed on heating
the metal in chlorine. Boiling concentrated hydrochloric acid slowly
dissolves lead : Pb + 2HC1 = PbCl2 + H2. Lead dichloride is
usually prepared, as a white precipitate, by adding a chloride to a
solution of a lead salt : Pb" + 2C1' ±^ PbCl2 (dissd.) ^±PbCl2 (ppd.)

In solution it appears to ionise in two stages : Pb012 ^± PbCl' +
f T -± Pb" -f 2C1'. The salt is sparingly soluble (1 per cent.) in cold
water, more soluble (3 -2 per cent.) in boiling water ; on cooling the boil-
ing solution anhydrous needles of PbCl2 separate. Lead chloride melts
at 485°, and boils at 956° ; the vapour density at 1070° corresponds
with PbCl2. It dissolves in concentrated hydrochloric acid, forming
hydrochloroplumbic acid, HPbCl3, salts of which are known. On
boiling litharge with a solution of common salt, partial decomposition
occurs, with formation of caustic soda (Scheele, 1773): 5PbO + H2O-f-
2NaCl ;=± 2NaOH + PbCL,,4PbO. The residue may be decomposed
by lime-water : PbCl2,4PbO + Ca(OH)2 = 5PbO + CaCl2 + H2O,
and the reactions have been used in the manufacture of alkali. On
heating the residue from the first reaction, a yellow lead oxychloride,
PbCl2,4PbO, known as Turner's yellow (1787), used as a pigment, is
formed. Another oxychloride, called Naples yellow, or Cassel
yellow, is prepared by heating litharge with ammonium chloride.

Lead fluoride, PbF2, and lead bromide, PbBr2, are formed by preci-
pitation. Lead iodide, PbI2. is formed as a yellow powder by pre-
cipitation. On boiling, it dissolves, and on cooling golden-yellow
spangles of the salt separate. It is soluble in a large excess of potass-
ium iodide, forming a double salt, decomposed on dilution. If
starch is added to the solution, it becomes blue on exposure to light,
indicating decomposition. Lead chlorate, Pb(C103)2,H20, is formed
from litharge and chloric acid. It evolves oxygen and chlorine on
heating.

If lead dioxide is dissolved in cold concentrated hydrochloric
acid and chlorine passed in, a dark brown solution is formed (Millon,
1842), containing hydrochloroplumbic acid, H2PbCl6. On adding
ammonium chloride to the solution, a yellow precipitate of ammonium
chloroplumbate, (NH4)2PbCl6, is formed. When this is added to cold
concentrated sulphuric acid, the free acid, H2PbCl6, which is probably
first formed, breaks up at once and a yellow liquid, which is lead
tetrachloride, or plumbic chloride, PbCl4, is deposited (Nilkoljukin,
1885). This has a sp. gr. of 3-18, freezes at — 15°, and readily
decomposes on warming, with evolution of chlorine : PbCl4 =
PbCl2 -|- C12. At 105° it explodes.

An orange-coloured solution of hydrochloroplumbic acid is formed
by the electrolysis of concentrated hydrochloric acid with a lead anode.

926 INORGANIC CHEMISTRY c HAI-.

By electrolysis of concentrated sulphuric acid with lead electrodes,
plumbic sulphate, Pb(SO4)2, is formed in yellow crystals. Lead tetra-
tiuoride is formed by the action of concentrated sulphuric acid on the
salt 3KF,HF,PbF4, obtained by fusing PbO2 with KF, and dissolving
in HF. This salt on heating first evolves HF, and then fluorine :
3KF,HF,PbF4 = 3KF,PbF4 + HF = 3KF + PbF2 + F2 + HF.

On the addition of a little water, PbCl4 forms a crystalline hydrate,
but it is readily hydrolysed, giving a brown precipitate of hydrated
lead dioxide. The ion Pb"", in fact, appears (like Sn"") to be very
unstable ; the insoluble dioxide is usually formed when the ion
might be expected.

Lead sulphide, PbS. — Lead burns in sulphur vapour, forming a
greyish-black mass of lead sulphide, PbS, which occurs as the mineral
galena. The sulphide is also formed as a black precipitate on passing
H2S into a solution of a lead salt. It dissolves in boiling dilute nitric
acid, with separation of sulphur ; the concentrated acid converts it
into a white mixture of PbS04 and Pb(N03)2. The sulphide melts at
1120°, and at higher temperatures sublimes. If H2S is passed into
a solution of a lead salt containing excess of hydrochloric acid, a
yellow or red precipitate is first formed, consisting of PbS,PbCl2.
This afterwards forms black PbS (cf. HgS; p. 876).

Lead pentasulphide, PbS5, is formed as an unstable purple precipitate
on adding a solution of CaS5 to a solution of a lead salt at 0°.

Lead nitrate, Pb(N03)2.— Lead nitrate is deposited in anhydrous
milky-white octahedral crystals, isomorphous with Ba(N03)2,
from a solution of lead, litharge, or lead carbonate in dilute nitric
acid (Libavius, " Alchymia," 1595). Concentrated nitric acid
precipitates it from aqueous solutions, and lead is not dissolved by
the concentrated acid because a protective coating of nitrate is
formed. On heating, lead nitrate evolves nitrogen dioxide (with
decrepitation) : the reaction if carried out in a sealed tube at
357° is reversible : 2Pb(N03)2 =±r 2PbO + 4N02 + O2. A basic
nitrate, Pb(OH)N03, is formed by boiling a solution of the nitrate
with litharge.

Lead sulphate, PbS04. — This salt is formed by adding sulphuric
acid to a soluble lead salt. It is a heavy white powder, difficultly
soluble in water (1 in 12,000), and almost insoluble in dilute sulphuric
acid (1 in 36,500). It dissolves in a warm solution of ammonium
acetate (BaSO4 is insoluble), or in hot concentrated sulphuric acid ;
on cooling the latter solution (6 per cent. PbSO4), crystals of PbS04,
H2S04,H2O are deposited. Lead sulphate occurs in crystals as
anglesite, usually isomorphous with barytes or celestine, but some-
times found as pseudomorphs of galena, and formed by the oxidation
of the latter. On boiling with ammonia, a basic sulphate, Pb2S05,
or 2PbO;SO3, is formed. PbS04?3PbO also exists.

XLIV THE METALS OF THE FOURTH GROUP !)27

Plumbic sulphate, Pb(SO4)2, may be formed in the accumulator :
PbSO4 -f SO4 = Pb(SO4)2 ; it is decomposed by water • Pb(SO4)2 +
2H2O = PbO2 H- 2H2SO4. It is a greenish-yellow substance produced
when sulphuric acid is electrolysed below 30° with a lead anode in a
porous pot.

Lead ehromate, PbCr04. — This compound is formed as a yellow
precipitate, insoluble in dilute, but soluble in concentrated nitric
acid (cf. BaO04), and is used as a pigment (chrome yellow). It is
probably the least soluble salt of lead.

Basic chromates, of orange or red colour, are obtained when the
normal ehromate is treated with boiling dilute alkali. Lead
ehromate is also precipitated when a lead salt is added to a solution
of potassium dichromate, but an equilibrium is set up unless an
acetate is added: K2Cr207 -f Pb(N03)2 — 2KN03 + PbCrO4 -f
Cr03. The acetate removes the chromic acid. PbCrO4 dissolves,
forming a yellow liquid, in concentrated caustic soda ; a complex
anion containing lead is produced : PbCr04 + 4NaOH — Na2Pb02
+ Na2Cr04 + 2H2O.

Lead cannot, therefore, be separated completely from acid radicals
in the ordinary process used in qualitative analysis, viz., boiling with
sodium carbonate, if a chromate is present. If the solution is reduced
with H2S, a chromic salt, and a precipitate of PbSO4 are produced.

Mixtures of lead chromate with lead sulphate or barium sulphate
are also used as yellow pigments. In calico-printing, the cloth is
mordanted with a lead salt, and then steeped in potassium chromate.
Lead chromate is used, instead of cupric oxide, in carrying out organic
combustion analyses (p. 691) when halogens are present. The lead
halides are non-volatile, whereas cupric chloride, etc., are volatile,
and pass over into the potash bulbs.

Lead phosphates, Pb3(P04)2, PbP207, and Pb(P03)2.— These com-
pounds are formed as white precipitates on adding the corresponding
sodium salts to a solution of lead nitrate or acetate. The orthophos-
phate dissolves in boiling phosphoric acid, and crystals of the acid
phosphate, PbHPO4 separate. The compound Pb3(PO4)2,Pb2Cl(PO4)
occurs as pyromorphite ; it is isomorphous with the mineral mimetite,
Pb3(AsO4)2,Pb2Cl(AsO4).

Lead borate, Pb2B6Ou,4H2O, used as a paint drier, is formed as a white
precipitate ; glassy borates are formed by fusing litharge with B2O3.
With the proportions 3PbO : 2B2O3 a soft yellow glass is formed ; with
4B2O3, a paler and harder glass is obtained, whilst with 6B2O3 a hard
colourless glass, of high refractive index, results. Lead silicates,
2PbO,SiO2, PbO,SiO2, and probably 3PbO,2SiO2, are similarly pro-
duced^ Mixtures of the borates and silicates form boro-silicate optical
glasses.

928 INORGANIC CHEMISTRY CHAP.

Lead acetate. — An important lead salt is the acetate,
Pb(C2H302)2,3H20,

known as sugar of lead on account of its sweet taste. It is prepared
by the solution of lead oxide (PbO) or carbonate in hot dilute acetic
acid followed by evaporation and crystallisation. Excess of lead
oxide must riot be added, otherwise a sparingly soluble basic salt is
formed. (This also occurs in the preparation of the nitrate). By
boiling litharge with a solution of lead acetate, a solution of a basic
acetate, called Goulard's extract, is formed, which is used as a lotion.
Two definite basic acetates are known : PbAc2,Pb(OH)2 and
PbAc2,2Pb(OH)2. By dissolving lead dioxide in warm acetic
anhydride lead t elm-acetate,., Pb(C.2H302)4, is formed, and separates
in stable white needles.

Lead carbonate, PbC03. — Solutions of lead salts give a white
crystalline precipitate of lead carbonate, PbC03 (sp. gr. 6-43). when
a solution of a carbonate is added in the cold. The precipitate is
sparingly soluble in water (1 in 50,500), but dissolves readily in a
solution of ammonium acetate. A complex anion appears to be
formed. The basic carbonate, 2PbC03,Pb(OH)2, is prepared as a
white pigment, known as white lead. Good white lead is an amor-
phous powder, consisting of globules 0 '0000 1-0 -00004 in. diameter ;
it mixes readily with linseed oil, and has a covering-power surpassed
only by lithopone. If improperly made, the substance is crystalline
and then has a considerable degree of transparency, its covering-
power being correspondingly reduced. White lead is readily black-
ened by sulphuretted hydrogen in the atmosphere. Its adultera-
tion by the cheaper barium sulphate is detected by the insolubility
of the latter in dilute nitric acid. Venetian white is a mixture of
equal parts of white lead and barium sulphate ; in Dutch white the
proportions are one to three.

The old Dutch process (described by the Latin Geber) produces a
white lead of excellent quality, but is tedious. Lead plates made into
rolls, or grids of cast lead, are placed in earthenware pots, with a per-
forated shelf at the bottom, and vinegar poured in below the shelf.
The pots are loosely closed with lead covers and stacked in rows, covered
with planks, interstratified with horse-dung or spent tan-bark, the
decomposition of which keeps the pots warm and produces carbon
dioxide. Basic lead acetate is probably first produced, and is then
decomposed by the carbon dioxide, the acetic acid set free again
entering into reaction :

(1) 2Pb + O, + 2H2O = 2Pb(OH)2 (in presence of air and moisture).

(2) Pb(OH). + 2CH3CO,H = Pb(CH,-CO,). + 2H0O.

(3) Pb(CH, C02), + H20 + C02 = PbC03 + 2CH3-CO,H.

(4) 2PbC03 4- Pb(OH)2 = 2PbC03,Pb(OH)2.

The plates after four or five weeks become encrusted with white
lead. This is stripped oft', washed, and ground. The moist paste is

XLIV THE METALS OF THE FOURTH GROUP <>:>9

dried in vacuum ovens or kneaded with linseed oil in pug-mills, when
water is displaced, the particles of white lead adhering strongly to
the oil.

By boiling litharge with lead acetate solution a basic acetate is
formed, which is precipitated by a stream of carbon dioxide. The
white lead made by this method ( Thenard's process) is, however, of
inferior quality.

A good product is obtained in the Dale and Milner process. Four
parts of lithargs are ground with 1 part of common salt and 16 parts
of water for fourteen hours, and carbon dioxide is passed in until the solu-
tion is neutral. The Bischof process also gives a good quality of white lead.
In this, litharge is reduced by water-gas at 250-300°, when the suboxide,
Pb2O, is formed. With water this gives the yellow hydroxide, PbOH,
which is converted by a current of carbon dioxide into white lead.

Germanium, Ge = 71-9. — The extremely rare element was dis-
covered by Winckler in 1886 in the mineral argyrodite, GeS2,4Ag2S.
The metal is easily reduced, and resembles tin and lead, but is brittle.
The oxides, GeO, GeO2, are known. The tetrachloride, GeCl4, and
germanium chloroform, GeHCl3, are stable compounds. Germanium
sulphide, GeS2, is white, and a gaseous hydride, GeH4, is obtained in an
impure state by adding a germanium salt to zinc and dilute sulphuric
acid.

Titanium, Ti = 47-72. — This element was discovered by Gregor
in 1798 in ilmenite, or titaniferous iron ore, which is ferrous titanate,
FeTiO3. The dioxide, TiO2, occurs in the minerals rutile, brookite,
and anatase, and in many rocks, clays, and iron ores. It is a white
powder used in forming a yellow glaze on porcelain, and in tinting
artificial teeth. Metallic titanium (containing carbon) is obtained
by reducing the dioxide with carbon in the electric furnace, or (in
the pure state) by heating the dioxide with calcium. An alloy
with iron, ferrotitanium, is prepared by reducing ilmenite with
carbon in the electric furnace, and is used in making special steels.
The tetrachloride, TiCl4, is a colourless, fuming liquid obtained by
heating the oxide with carbon in a current of chlorine ; it is partly
hydrolysed by water. The solution is reduced by zinc and hydro-
chloric acid to a deep violet trichloride, TiCl3, which is a powerful
reducing agent. Hydrogen peroxide gives with titanium salts a
bright yellow colour, due to the trioxide, Ti03.

Zirconium, Zr = 89-9. — The mineral zircon occurs in alluvial
sands in Ceylon and in other localities, and consists of zirconium
silicate, ZrSiO4. From this zirconium dioxide, Zr02, or zirconia,
was obtained by Klaproth in 1789. Zirconia is used as a refractory,
and (mixed with rare earths) in forming the filaments of Nernst
lamps, which become conducting on heating. The metal is obtained

3 o

930 INORGANIC CHEMISTRY CH. XLIV

by reducing a fluozirconate, K2ZrF6 (cf. K2SiFe), with potassium,
or in the electric furnace. When alloyed with iron it forms a very
tough steel.

Thorium, Th = 230-31. — Thorium occurs in the minerals thorite
(chiefly thorium silicate), thorianite (mainly thoria, ThO2), and mon-
azite, a phosphate of cerium and lanthanum containing 4-18 per
cent, of thoria. Monazite occurs in the form of alluvial sand in India
and Brazil. Thorium compounds are used in the manufacture of
Welsbach incandescent gas mantles, which consist of cellulose im-
pregnated with a mixture of thorium and cerium nitrates, which on
ignition leaves a mixture of 99 parts of thoria and 1 of ceria. Pure
thoria emits a relatively feeble light. A peroxide, Th207, is precipi-
tated by alkaline H202. Thoria is also added in small amounts
to tungsten electric lamp filaments (p. 958) : it prevents disin-
tegration of the latter in use.

EXERCISES ON CHAPTER XLIV

1. What are the common ores of lead and tin ? How are the metals
obtained from these ores ? By what tests would you distinguish lead
from tin ?

2. What oxides of lead are known ? How are they prepared ?
Describe briefly the properties of these oxides, with special reference to
their acidic or basic character.

3. How are the higher chlorides of lead and tin obtained ? What is
the action of acids on these metals ?

4. Describe the manufacture of white lead. What substitutes have
been proposed for white lead as a paint, and for what reasons ?

5. By what experiments would you demonstrate that stannous salts
are reducing agents ? Give equations.

6. By what reactions is tin separated from arsenic in qualitative
analysis ? How are the sulphides of tin obtained ?

7. How are the stannic acids and stannates formed ? Do any
corresponding lead compounds exist ?

8. Describe the reactions taking place in the charging and discharging
of a lead accumulator.

CHAPTER XLV

THE METALS OF THE NITROGEN GROUP

The metals of the nitrogen group. — The fifth group in the periodic
table includes, besides nitrogen, phosphorus, and arsenic, a number
of metals, all of which (except antimony and bismuth) are rare.
The group is divided into two sub-groups, as follows :

Even series. Odd series.

Vanadium, V = 50-6 Nitrogen, N = 13-897

Niobium, Nb = 92-4 Phosphorus, P = 30-79

Tantalum, Ta =180-1 Arsenic, As = 74-37

Antimony, Sb = 119-2
Bismuth, Bi = 206-4

The element niobium is sometimes called columbium, Cb. The
members of the two sub-groups resemble one another very closely
in chemical properties, but differ in some respects. One important
difference, which indicates that the division into odd and even
series indicated by the periodic classification is not merely arbitrary,
is that the members of the even series do not form organo-metallic
compounds with hydrocarbon radicals, whilst the elements of the
odd series form stable compounds of this character. This difference
is found throughout the periodic system : the elements of even
series do not form organo-metallic compounds except in Group VIII.

All these elements form typical acidic oxides. R20s> the acidic
character diminishing with increasing atomic weight. The metals
vanadium, niobium, and tantalum, in the even series, combine very
readily with oxygen, and their compounds are extremely difficult
to reduce. They have high melting- and boiling-points and a metallic
appearance. The elements of the odd series, on the contrary, are
easily reduced from their compounds, have low melting points,
and are readily volatilised. In the odd series the gradual tran-
sition from typical non-metals to typical metals is very clearly exhibited.
The element phosphorus is decidedly a non-metal, but antimony and
bismuth are typical metals, although they are brittle. Arsenic,
which stands on the threshold between the two classes, is sometimes
regarded as a metal, sometimes as a non-metal ; it shows properties
belonging to both groups of elements. Elements of this kind are
sometimes called metalloids.

931 3 O 2

932 INORGANIC CHEMISTRY CHAP.

Compounds of the two types RX3 and RX5 are formed by all
the elements of this group ; in addition, some compounds in which
the element is bi- and quadri-valent are known. Thus, nitrogen is
bivalent in NO ; vanadium forms a dichloride, VC12, and a tetra-
chloride, VC14.

ANTIMONY, Sb = 119-2.

Stibnite. — The very earliest records mention under various names
a substance used as a pigment, and for painting the eyebrows and
face. This latter practice appears to date from prehistoric times ;
it was used in Egypt at least as early as 3000 B.C. The black pig-
ment came from Arabia, and was called stimmi, afterwards stibi.
The substance was native antimony sulphide, Sb2S3, now known as
stibnite. In II. Kings ix. 30, we find in the translation of St.
Jerome : " Porro Jezebel introitu ejus audito pinxit oculos suos
stibio," the last word being a literal translation from the Hebrew
for stibnite.

Metallic antimony is very easily reduced from stibnite, and a
Chaldean vase of date 3000 B.C. was found by Berthelot to consist
of pure metallic antimony. The metal, however, was not speci-
fically referred to by ancient writers, and was probably confused
with lead. Constantinus Africanus (c. A.D. 1050) refers to stibnite
as antimonium, and the metal was well known to the alchemists.
The preparation of metallic antimony and of a number of its com-
pounds is clearly described by Basil Valentine (or Threlde, see
p. 29) in the " Triumphal Chariot of Antimony," Leipzig, 1604,
and antimonial compounds had been extensively used in medicine
by Paracelsus. The Arabic name for finely-powdered stibnite, al
kohol, was applied by Paracelsus to the " quintessence," and thence
to spirit of wine — alcohol.

Metallic antimony. — In the preparation of metallic antimony,
stibnite, which occurs abundantly near Oporto, is first liquated,
i.e., heated so that the readily fusible sulphide of antimony (m.-pt.
540°) flows away from the rocky portion. The sulphide is then
reduced by heating with iron and a little salt in plumbago crucibles :
Sb2S3 -f 3Fe = 2Sb + 3FeS. The metal (regulus of antimony)
melts, and collects below the slag. The sulphide may also be
carefully roasted in a reverberatory furnace, when, at 350°, antimony
dioxide, Sb2O4, is left. At higher temperatures, the trioxide, Sb203
(or Sb4O6), sublimes, the arsenic volatilising first as trioxide :
2Sb2S3 + 902 = Sb406 + 6S02. The antimony oxides are mixed
with charcoal and sodium carbonate and heated to redness, when
reduction occurs : Sb406 + 60 = 4Sb + 6CO. The regulus is
purified by fusing with sodium carbonate and a little nitre. It
then crystallises on cooling in beautiful star-shaped forms, men-
tioned by Basil Valentine. The total production of antimony in

XLV THE METALS OF THE NITROGEN GKOll' 933

1912 was estimated at 35,000 tons, the greater proportion being
supplied by France.

Pure antimony is prepared by fusing the pentoxide, prepared by the
hydrolysis of recrystallised chlorantimonic acid (p. 937), with potassium
cyanide.

Properties of antimony. — Antimony is a silver-white, lustrous
metal, sp. gr. 6-8, which is brittle and easily powdered. From the
fused metal, on slow cooling, large obtuse rhombohedral crystals
are formed, but after rapid cooling the metal has a granular struc-
ture. Antimony melts at 630-5°, and boils at 1440°. The
vapour densities at 1572° and 1640° correspond with the molecular
weights 310 and 284, respectively. These are intermediate between
Sb3 and Sb4 ; the vapour possibly consists of Sb4 -}- Sb, which
would correspond with a density of J(120 -f- 480) = 300. The
metal is precipitated as a fine black powder when zinc is added to a
solution of the trichloride ; this powder is used in covering plaster
casts to give them the appearance of steel.

Antimony is unchanged in air, and is not acted upon by water or
dilute acids. It decomposes steam at a red heat, and is oxidised
by concentrated nitric acid, giving oxides of nitrogen and a white
powder of antimonic acid. Antimony dissolves readily in aqua
regia, forming a solution of the pentacbloride, SbCl5.

When heated in air antimony burns, evolving white fumes of
the trioxide, Sb203, and tetroxide, Sb204. A bead of antimony
heated on charcoal before the blowpipe continues to burn when
the flame is removed : if dropped on a piece of paper turned up at the
edges, the bead breaks up into burning globules, which disperse and
leave curious charred tracks on the paper.

A 11 o tropic forms of antimony. — Unstable allotropic forms of antimony
are known. Yellow, or a- antimony is produced by the action of
ozonised oxygen on liquid stibine, SbH3 (q.v.) at — 90°. It is amor-
phous, and is slightly soluble in carbon disulphide. Yellow antimony
is very unstable, and passes readily at temperatures above — 90° into
black antimony, an amorphous black powder, sp. gr. 5-3, which is
formed directly from liquid stibine and oxygen at — 40°. Black
antimony oxidises spontaneously in air, and on warming forms ordinary
rhombohedral, or ,4 -antimony with evolution of heat. Amorphous
antimony was obtained by Gore (1858) by the slow electrolysis of a
concentrated solution of the trichloride in hydrochloric acid with a
platinum cathode and an antimony anode. The metal deposited
on the cathode resembles polished graphite, and has a density of 5-78.
When scratched, it falls to powder with a slight explosion, evolving
fumes of SbCl3, which it always contains to the extent of 4-12 per cent.
At 200°, it explodes violently. Amorphous antimony can be kept under

'.»:u INOKC xxir riiKMisrm < MM-.

\xator. luil it i ho Inllor ib hentod to T;> the (intimom explode*. This
form is probably t\ solid solution of Shri3 in blavk antimony

Alloys Of antimony. Antimony is a eonstituont of sex oral
important alloys. A mixture of 1 ."> parts of antimoin and S."> of
lead is hunt Iciiil. or tintitnoHial /<W. used for stopeoeks for snlphurie
••»» 'iii The most important alloxs of Iho motal nro tlu>so uso»l fi>r
printers' typ<\ Tlio (\-irly printers used MOOO!(MI type: ??»r/n7 /t/;>r
v-ontainm;: antimony is refc^rnnl to by Kasil \'al(M\tiiu> as in common
use in 1 <)<><>.

Anti- Bis-

Lead. mony. Tin. Copper. Zinc. muth.

r\pr moi ... <;o :?o 10

Linotypo inolal ... S:<-:>

Monot\|x> inotMl ... SO 1 .">

Hritjiniu:i motal ... — 10 I M<1 10

IVwtor ...... — 7-1 SO-S IS IS

Anli fn.MitMi Ix'jvrini;

niotnl ...... <tO 20 20

Tho liolinito rompoiiMtl. Sl>( u ,. is a houutiful purplo :vllo> .

Oxides Of antimony.- Antinu>n\ forms t\\o series ol vompounds,
sl>\ ami Sl>\ , In solution thoso appi>ar to jjivo tlu^ ions Sl>
and Sl>"-: . althougli hydrolysis otvurs to a larvr»* extent. Three
oxides are kno\\n

Antimony trioride, Sl>.2()., (or Sl>4()6).
Antimony tetroxide, Sb,(), (or Sb()2).
Antimony pentoxide, SI

Tlu'se are all acidic oxides, although tlu^ trioxule also sh»n\s
\v<\ikl\ l»a,sie properties. It dissoKes in e»>ld eoneent rated nitrie
aeid. forming antimony nitrate, Sl>(N(V,V,. in hot eoneeiitrat<^d sul
phurie aeid, forming antimony sulphate, SbfSO.).^ and rendil\ m
• lilute livdroeldtMMe aeid t-o form the trichloride, SbCIg, or in t-artarie
aeid. All tlie oxides are easilx nxluee«l by hydrogtMi or earboi\

Antimony trioxide, 8b2O.,. oeeurs natixe as scmmmwitc in enbii-
erxstals. and more rarely in rhombie erystals as nricntinitf. It is
obtained as a pah1 bntf eoloured poxx der bx digesting antimony
oxvrhloride. Sb(XM, xxith a solution of sodium earbonate. or by
passing steam oyer red hot antimony. 1'Yom a hot solution in
sodium earbonate both forms are deposited in xxhite erxstals
. \ntuuonx trioxide hoeomes yelloxx on heating, being apparently
eonxrrted int»> th(^ rhombie form, but heroines xxhite on eooiing.
It fust^s at a n»d beat, and xolatilises at \M^ . \\\c xa}>our density
I'orn^spt^nding xxith Sh ,<>,.. The trii>xide disst>lxes in alkalies,
forming salts, e.g.. XaSbO.,. :U1.:(>. domed from a hxpothetieal
metantimonions acid. HShO., The sodium salt is sparingly soluble

TIM MII \i s MI i in M i i;,M,r\ ,,I;,H i- <»:;.,

m water and crystallises m glittering octahcdra The potassium
salt, Ix ,<>,:'»Sh.O.', obtained by fusing Sb <>. mil, potash, is readily
soluble m \\ater I! dilute nitric or sulphuric acid i i added to larlar
emetic (»/ r.) the precipitate \\hen dried al 1OO ha . the composition
H.ShO.. ortluwntimonious acid. Pyroantinionious arid, ll,Sl>.<> i
obtained as a \\lnte precipitate by addini1 copper sulplial

solution of antimony trisulphide m caustic
prccipiiale h. i-ms lo lorm (at tirst a yellow
do\\n) and then adding acetic acid, I'M
the trioxide

Antimony tetroxulr Sb,(),, is ..blamed

heal Hi!1 t he I rn >x ide m air a I .'I'.M > 7 ,'.»

it decompose., into Sb .< ) . Tin- pentoxi.

telroxide on lie.itiniv Impure Sb ,( ) , is .

mte ; if the oxidation 18 incomplete, the

f//(/,s',s of ( tnlini <>//?/. l< consists of tetroxule with unchauj

and is used in colouring glass and porcelain vcllow

teti'oxide form all i when fused with alkalies, knonn

ttionittlcn. If the fused mass obtained from Sb,()1 and Ix ( H I

boiled with \\ater and the solution precipitated with hydrochloric

acid, the salt K..Sb,( ),,, or Ix ,< >,:'SI>,< ),, is obtained

Antimony pentoxido and antimonlates. Antimony pentoxlde,

Sb () is obtained as a yellon po\\der b\r gentl\' healiii". the solid
produced by the repeated evaporation of antimony \\iih conceit
(rated nitric acid At -HO" it decomposes rapidly into Sb..t>,. and
when prepared as de .cnbed al\\a\.. contain, a little loner oxide

When antimony i, fused \\ith pota.sium mlrati -, and the re. idiie
exlracled \\ith cold \\ater, a \\lute ponder of potiiHSinm uiclaiili
uioniate, KSM).t, remains, which is soluble in boiling \vat.er. Hilut(>
nitric acid piv< ap:l alea from the .olulioii a hydrated penloxide,
which < >n gentle heating bn'in. an(imon\ pentoxide, Sl>,<) m a
pure state. This redden, moist lilmu paper although it is prae
tieall\ insoluble m \\aler A h\drated form ol Sb ,( ),t is also formed
b\ oxidi. in:1 tin- irioxide m presence of \\alcr \\ith iodine, chlorine,
or potassium dichromatc. \\ith bromine, nitric acid, <>r a mixture
of pola.siiim chlorate and h\ drochloric acid the oxidation is
incixuplete \\ hen antimony pent achloride (</ r ) i precipitated
\\ilh hot \\aler. or the trichloride or one of the loner oxides treated
with nitric acid, the residue after wa,,hiii" and lieahii" lo H.n'
corresponds in composition with pyrouutimonic acid, II .jSbJ )7.
At LM»n , (hi, i., said to form mctuiitiinoiiii: and, MSbO.,. Ortbo-
antimonic ncid, ll.,Sb(),, is aid to be formed by precipitating
potassium antimoniale with dilute mine at id and <h \ ni" over
sulphuric acid in a desiccator.

INroanlimomc a. id dissolves in can itie pot i .h forming a gummy
\\lueh deposits potassium met ant imoniate, KSbO,, in Crystals

936 INORGANIC CHEMISTRY CHAP.

up to a certain point, but on further evaporation yields a gummy
mass. The solution forms with sodium salts a white, amorphous
precipitate, possibly NaSbO3, which rapidly becomes crystalline
and then consists of acid sodium pyroantimoniate, Na2H2Sb207,6H20,
sparingly soluble in cold water (1 in 350), and almost insoluble in
alcohol. This is the least soluble sodium salt, and a solution of
potassium metantimoniate (obtained from antimony and nitre as
described) may be used as a test for sodium salts. A corresponding
acid potassium pyroantimoniate, K2H2Sb2O7,6H2O, is obtained by
oxidising potassium antimonite with potassium permanganate.
Ammonium metantimoniate, NH4SbO3. is obtained in crystals from
a solution of antimonic acid in warm ammonia.

Halogen compounds of antimony. — Halogen compounds of types
SbX3 and SbX5 are known. Those of type SbX3 are known with
all. the halogens ; SbX5 occurs only as SbF5 and SbCl5.

SbF3, white solid. SbF5, viscous liquid, b.-pt. 155°.

SbCl3, white soft crystals, m.-pt. SbCls, yellow mobile liquid, b.-pt.

73-2°, b.-pt. 223-5°. 140°.

SbBr3, white deliquescent needles,

m.-pt. 95°, b.-pt. 275°.
SbI3, three forms, dark-red and

greenish - yellow. M.-pt of

stable form 171°.

Antimony trichloride, SbCla. — This compound was prepared by
«Basil Valentine by distilling roasted stibnite with corrosive sub-
limate : Glauber (1648) obtained it by dissolving stibnite in hot
concentrated hydrochloric acid : Sb2S3 -f 6HC1 = 2SbCl3 + 3H2S.
The dark brown solution is distilled ; water first passes over, then
hydrochloric acid, and finally antimony trichloride, which solidifies
in the receiver as a white, crystalline mass (butter of antimony}.

Antimony trichloride is decomposed by water, with deposition
of white basic chlorides. It forms a clear solution with hydrochloric
acid, from which crystals of chlorantimonious acid, 2SbCl3,HCl,2H20,
may be obtained. In solution, the compound is probably H3SbCl6,
stable salts of which, R3SbCl6, are formed with metallic chlorides.
The vapour density of the trichloride, and the boiling point of its
ethereal solution, correspond with SbCl3.

Antimonious oxychloride, is precipitated as a white powder when
a solution of the trichloride in hydrochloric acid is poured into
water. The composition of the precipitate, known as powder of
Algaroth, varies with the dilution. Two definite oxy chlorides are
known :

SbCl3 + H20 — SbOCl -f 2HC1 (formed with a little water) ;
4SbCl3+5H2Oz±Sb4O5Cl24-10HCl (with a larger amount of water).

XLV THE METALS OF THE NITROGEN GROUP !)37

With excess of water, especially on heating, hydrated antimony
trioxide is formed.

Antimony pentaehloride, SbCl5. — This compound is formed by the
action of excess of chlorine on the trichloride, or by treating the
latter with aqua regia. It is a heavy, yellow, fuming liquid, solidi-
fying on cooling (m.-pt. — 6°). The vapour is slightly dissociated
at the boiling point, 140° : SbCl5 ±^SbCls -f C12, but the compound
volatilises unchanged at 79° under 22 mm. pressure; the vapour
density corresponds with SbCl5 With ice-cold water, two crystalline
hydrates, SbCl5,H2O (soluble in chloroform) and SbCl5,4H2O
(insoluble in chloroform), are formed. With excess of water,
antimonic acid is produced. When the pentaehloride and pentoxide
of antimony are heated in the proportion 3SbCl5 : Sb205 at 140°,
two oxychlorides, Sb3OCl13 (white, deliquescent crystals, m.-pt. 85°)
and Sb3O4Cl7 (yellowish-white crystals, m.-pt. 97-5°) are formed.
With concentrated hydrochloric acid, a fairly stable crystalline
chlorantimonic acid, 2HSbCl6,9H20, is formed. This may also be
prepared by passing chlorine through a solution of the trichloride
in hydrochloric acid, and then adding excess of concentrated hydro-
chloric acid.

The brown liquid obtained by the action of chlorine on SbCls appears
to contain a tetrachloride, SbCl4, or H2SbCl6 ; stable salts of dark
colour, e.g., Rb2SbCl6, are known.

Antimony trifluoride, SbF3, is obtained by distilling antimony with
mercuric fluoride, or by dissolving the trioxide in hydrofluoric acid and
evaporating. It is not hydrolysed by water. Potassium fluoranti-
monite, K2SbF5, prepared by dissolving Sb2O3 in a solution of KF in
HF, is used in calico-printing. The tribromide and tri-iodide are formed
from the elements ; they are decomposed by water, yielding SbOBr
and SbOI. The vapour of SbI8 is scarlet in colour. Antimony penta-
fluoride, SbF5, is a colourless, oily liquid, without action on glass when
dry, obtained by boiling the pentaehloride with anhydrous hydro-
fluoric acid under a reflux condenser for three days and fractionating.
The apparatus must be constructed of platinum.

Sulphides of antimony. — Two sulphides of antimony, SbaS? and
Sb2S5, are known. The trisulphide, Sb2S3, occurs as the grey mineral
stibnite, sp. gr. 4*652. By precipitating a solution of antimony
trichloride in hydrochloric acid with sulphuretted hydrogen, a
red, amorphous precipitate is formed, which if dried at 100° and then
heated in carbon dioxide to 200° forms the greyish-black modifica-
tion. The red form is used as a pigment (antimony vermilion), and
in vulcanising rubber, the red varieties of which contain it. If the
black form is heated at 850° in a stream of nitrogen, and the vapour
rapidly cooled, lilac-coloured globules of a third form, sp. gr. 4-278,
are formed. The red precipitate is insoluble in dilute acids, but

938 INORGANIC CHEMISTRY CHAP.

dissolves in hot concentrated hydrochloric acid. If the solution,
still containing H2S, is diluted with water, red Sb2S3 is precipitated.
Colloidal Sb2S3 is formed as an orange-red liquid by adding a
0-5 per cent, solution of tartar emetic to sulphuretted hydrogen
water.

Antimony trisulphide is reduced on heating in hydrogen ; the
reaction is reversible : Sb2S3 -f 3H2 ^± 2Sb -f- 3H2S. It is used,
mixed with nitre and sulphur, in the preparation of blue- fire in
pyrotechny, and in making matches (p. 626). It dissolves in alkali
sulphides, and hot concentrated solutions of alkalies and their
carbonates. On dilution, a red mixture of Sb203 and Sb2S3 (Kermes
mineral) is formed. The solutions, and the substances obtained on
fusion of Sb2S3 with Na2S, probably contain thioantimonites,
R3SbS3, R4Sb2S5, RSbS2.

Precipitated antimony trisulphide is insoluble in ammonium
carbonate, whereas arsenic sulphide is soluble (p. 656). It dissolves
in fairly concentrated hydrochloric acid on boiling ; arsenic sulphide
is insoluble.

Thioantimoniates. — When antimony trisulphide is boiled with
caustic soda and sulphur, the filtered and cooled solution deposits
pale yellow crystals of Schlippe's salt, or sodium thioantimoniate,
Na3SbS4,9H20. The compounds K3SbS4,9H20, (NH4)3SbS4, and
Ba3(SbS4)2,6H20 are also known. A solution of the ammonium
salt is obtained on dissolving the trisulphide in yellow ammonium
sulphide : Sb2S3 + 3(NH4)2S2 + 2S = 2(NH4)3SbS4. These salts
correspond with an unknown ortho-thioantimonic acid, H3SbS4, or
Sb2S5,3H2S ; on acidification, the acid is not produced but its thio-
anhydride (i.e., thio-acid — H2S), antimony pentasulphide, Sb2S5,
is precipitated :

2(NH4)3SbS4 + 6HC1 = 6NH4C1 -f Sb2S5 + 3H2S.

This forms a dark orange-red precipitate, mentioned by Basil
Valentine, and by Glauber (1654). On heating alone, or with water
or acids, it decomposes into sulphur and the black trisulphide.
Antimony pentasulphide readily dissolves in alkalies, even ammonia,
and alkali sulphides, forming thioantimoniates :

3Na2S = 2Na3SbS4 ;
4Sb2S5 + 24KOH = 5K3SbS4 + 3K3Sb04 -f 12H2O.

Antimony hydride, SbH3. — Antimony trihydride, antimoniuretted
hydrogen, or stibine, SbH3, is formed mixed with hydrogen, when a
solution of an antimony salt is added to zinc and dilute sulphuric
acid. The gas evolved burns with a white, luminous flame, pro-
ducing fumes of the trioxide. A black stain of antimony is deposited
on a cold porcelain dish held in the flame : 2SbH3 ^ 2Sb -J- 3H2.
This is also formed, on both sides of the heated spot, on passing

XLV

THE METALS OF THE NITROGEN GROUP

939

the gas through a heated glass tube (As is deposited from AsH3 only
on the side furthest from the generating flask).

To distinguish the product from the similar but brighter arsenic
mirror, three spots are formed on the dish, which are treated as follows :

( 1 ) Moisten with a

(2) Moisten with a

(3) Moisten with

solution of bleaching

concentrated solu-

yellow ammonium sul-

powder :

tion of tartaric acid :

phide, and evaporate :

As dissolves :

As is insoluble.

As gives a yellow

5Ca(OCl)2 + 6H2O -f

residue of As2S3.

As4 = 5CaCl2 4

4H3AsO4.

Sb is insoluble.

Sb dissolves, forming

Sb gives an orange

(SbO)2C4H4O6. residue of Sb2S3.

Pure stibine is prepared (Stock and Guttmann, 1904) by the action
of hydrochloric acid on an alloy of magnesium with 33 per cent, of
antimony : Mg?Sb2 + 6HC1 = 3MgCl2 -f 2SbH3. The gas is washed
with water, dried with calcium chloride, and passed into a tube
surrounded with liquid air. White solid stibine is formed, which
melts at — 88° to a colourless liquid boiling at — 17°. The gas
may be collected over mercury, and is fairly stable when dry. It
has an unpleasant odour, and is poisonous. It is attacked by air or
oxygen, forming water and antimony, and decomposes into its
elements in presence of moisture, or with explosion when heated or
sparked. The density is slightly higher than that corresponding
with the formula SbH3.

When hydrogen containing stibine is passed into a solution of
silver nitrate, a black precipitate is formed, and the filtrate contains
no antimony, whereas if arsenic hydride is present, the filtrate
contains the whole of the arsenic (p. 649).

The precipitate first formed is silver antimonide, SbAg3, but this
is rapidly decomposed by the excess of silver nitrate, forming a black
mixture of silver, antimonious acid, and a little antimony. If this
is warmed with hydrochloric acid, the filtrate gives with H2S an
orange-red precipitate of Sb2S3.

Estimation of antimony. — Antimony is estimated by precipitation
as sulphide, Sb2S3, which is heated in a porcelain boat in a stream
of carbon dioxide. The trioxide may be dissolved in tartaric acid,
neutralised with sodium carbonate, and titrated with iodine :
Sb2O3 4- 2H2O -f- 2I2 = 4HI 4- Sb205 ; or a solution in hydro-
chloric acid may be titrated with sodium bromate :

3SbCl3 4- 6HC1 4- NaBr03 = 3SbCl5 4- 3H20 4- NaBr.
Antimony pentoxide may be estimated by the reaction :

Sb2O5 4- 4KI 4- 10HC1 = 2SbCl8 4- 4KC1 + 2I2 4- 5H2O.

940 INORGANIC CHEMISTRY

The atomic weight of antimony has been determined by heatii
tartar emetic in hydrogen chloride and weighing the residual
potassium chloride : Sb = 119-2 (H = 1), or 120-2 (0 = 16).

Tartar emetic is an important medicinal preparation, obtained
boiling oxide of antimony with water and cream of tartar (potass-
ium hydrogen tartrate). It contains the radical antimonyl, SbO,
and is potassium antimonyl tartrate, 2K(SbO)C4H4Of5,H2O. It is
also used as a mordant.

BISMUTH. Bi = 2064.

Bismuth. — Metallic bismuth was probably known to the
ancients, but was confused with tin and lead. Agricola
(1546) describes it under the name of bisemutum, or plumbum
cinereum, noting that it was used to soften tin. The name is
supposed to have been derived from the German idismuth (a meadow),
given to it by the old miners on account of its red colour. Pott
(1739), and later Bergman, investigated its compounds, some of
which had been used by Paracelsus for medicinal purposes. The
basic nitrate — " bismuth subnitrate," Bi(OH)2NO3, discovered by
Libavius, is still used medicinally in diarrhoea and cholera. This
substance, known as pearl white, was introduced by Lemery as a
cosmetic, and still finds supporters, although it no doubt acts in-
juriously on the skin by reason of its slight hydrolysis in contact
with perspiration, with production of nitric acid. Lemery remarks
that its use is injurious.

Bismuth occurs somewhat sparingly, usually in the native con-
dition containing arsenic and tellurium, in Bolivia, Saxony, and
Australia. The oxide, Bi2O3. also occurs, as bismuthite, or bismuth
ochre, but the sulphide, Bi2S3, bismuthine, or bismuth glance, is rare.

The metal is obtained from native bismuth by liquation, the ore
being heated in sloping iron tubes, when bismuth, which has a low
melting point (271°) flows away. The oxide and sulphide ores,
which usually contain cobalt and nickel, are first roasted, when the
trioxide, Bi2O3, is formed. This is heated with charcoal, iron, and
a flux, and melted in crucibles or in a reverberatory furnace, when
metallic bismuth fuses, and collects below the cobalt and nickel
arsenides. The crude bismuth is purified by dissolving in dilute
nitric acid, pouring the solution of bismuth nitrate, Bi(N03)3, into
water, calcining the basic nitrate precipitated, and reducing the
oxide as before. Very pure bismuth is obtained by reducing the
pure oxide with potassium cyanide. The pure oxide is obtained by
heating the nitrate, which has been crystallised from a solution
containing a large excess of concentrated nitric acid.

Properties of bismuth. — Bismuth is a white metal, sp. gr. 9-78,
with a distinctly reddish tinge ; it readily forms large crystals on
cooling. These crystals, which are obtuse rhombohedra resembling

XLV THE METALS OF THE NITROGEN GROUP 941

cubes, are usually covered with a superficial film of oxide, and then
exhibit a splendid iridescent play of colours. The metal is brittle,
and is easily powdered. Bismuth and its alloys with other metals,
which have very low melting points, expand when they" solidify,
and the alloys are used as stereo -metal in printing, the cast being
made just before solidification.

Wood's fusible metal (m.-pt. 71°) contains 4 bismuth + 2 lead +
1 tin -f- 1 cadmium ; Rose's metal (m.-pt. 93-75°) contains 2 bismuth +
1 lead -f- 1 tin, and Lipowitz alloy (m.-pt. 60°) consists of 15 bismuth +
8 lead -j- 4 tin -j- 3 cadmium. Alloys of lead, bismuth, and tin, melting
slightly above 100°, are used in the construction of automatic sprinklers,
which discharge a spray of water over combustible goods in warehouses
when the fusible metal plug is melted by the rise in temperature resulting
from a fire. Fusible solder, which can be applied under hot water
containing a little hydrochloric acid, also contains the same materials.
Less fusible alloys are used as safety plugs in boilers.

Bismuth boils at 1400°, and the vapour density between 1600°
and 1700° shows that partial dissociation occurs: Bi2±^:2Bi.
This is complete at 2000°. The metal is unchanged in dry air,
and is only slowly attacked by water. When fused, however, it
is slowly oxidised to Bi203, and" when strongly heated burns with a
bluish-white flame, forming brown fumes of Bi203. It decomposes
steam slowly, liberating hydrogen. It is not attacked by dilute
acids in the absence of oxygen, with the exception of nitric acid,
which dissolves it, forming the nitrate, Bi(N03)3. It also readily
dissolves in aqua regia, forming the trichloride, BiCl3. Boiling con-
centrated sulphuric acid converts it into the sulphate, Bi2(S04)3,
sulphur dioxide being evolved. A colloidal solution of the metal is
formed by reducing the oxychloride with hypophosphorous acid.

Solutions of bismuth salts contain the ion, Bi"* , but they are
partly hydrolysed by water, producing precipitates of basic
salts ; these redissolve when an excess of acid is added, the reaction
being reversible (cf. PC13) :

BiCl3 + 2H20 — Bi(OH)2Cl + 2HG1 = BiOCl + H2O -f 2HC1.

Bismuth nitrate, Bi(N03)3. — The most important bismuth salt
is the nitrate, obtained in triclinic crystals, Bi(N03)3,5H20, by
evaporating a solution of the metal in warm 20 per cent, nitric acid.
A solution of bismuth in dilute nitric acid, if poured into a large
volume of water, deposits the white basic nitrate, or " subnitrate/'
Bi(OH)2N03. If this is repeatedly washed with water, white
bismuthous hydroxide, Bi(OH)3, is left. The hydroxide is also
precipitated by alkalies from the solution of the nitrate ; it is
insoluble in excess of alkali unless glycerol is added, but is readily
soluble in acids. When heated to 100° it forms BiO(OH), and
on ignition leaves a yellowish residue of bismuth trioxide, Bi2O3.

942 INORGANIC CHEMISTRY CHAP.

If crystals of bismuth nitrate are triturated with mannitol, the
mixture gives a clear solution with water. The pure salt can be
obtained in solution only if dilute nitric acid is added.

The compounds BiO(OH), BiOCl, and Bi(OH)2N03 contain the
univalent bismuthyl radical,. BiO-, corresponding with SbO-. The
basic salts obtained by adding a solution of a bismuth salt to water
are readily distinguished from the antimony salts by adding a few
crystals of tartaric acid and warming. The antimony salts dissolve,
but the bismuth salts are insoluble.

Oxides of bismuth. — The following oxides of bismuth have been
described :

Bismuth dioxide, Bi2O2 (feebly basic).
Bismuth trioxide, Bi2O3 (basic).
Bismuth tetroxide, Bi2O4 (acidic).
Bismuth pentoxide, Bi205 (acidic).

Bismuth trioxide, Bi203, which is obtained by heating the
hydroxide, BiO(OH), or directly by heating bismuth nitrate, is a
yellowish-white powder which fuses at 820°. On heating to 704°,
the powder changes with evolution of heat into a second form, con-
sisting of greenish -yellow crystals. A third form is obtained in
yellow needles by heating the oxide in a porcelain crucible to the
melting point. Bismuth trioxide is used in producing an iridescent
white glaze on porcelain. When mixed with other oxides and fused
on the surface of glass, it is used in making stained glass. Thus,
with chromium sesquioxide, a lemon-yellow tint is obtained.

Bismuth dioxide, Bi202, is formed as a black powder on heating
the basic oxalate : (BiO)2C204 = Bi202 -f 2C02. It is precipitated
on adding a solution of 1 part of stannous chloride to 1 part of
bismuth trioxide suspended in caustic potash solution. The black
precipitate is washed with dilute potash solution and dried at 100°.
Bismuth dioxide burns when heated in air, forming Bi203. A pre-
cipitate consisting of the black dioxide mixed with white bismuth
hydroxide or stannous hydroxide is obtained when a solution of a
bismuth salt is added to a solution of stannous chloride in excess of
caustic soda solution. The formation of a mixture of black and white
precipitates so obtained constitutes the magpie test for bismuth or tin.

When bismuth trioxide is oxidised with alkaline potassium
ferricyanide solution, a brown powder of bismuth tetroxide, Bi204,
is obtained. By passing chlorine into a suspension of bismuth
trioxide in nearly boiling caustic potash, a scarlet powder of bismuth
pentoxide, Bi205, is formed. Both these resemble lead dioxide in
colour, and in being insoluble in nitric acid :

Bi203 + 2K3FeC6N6 + 2KOH = Bi204 + 2K4FeC6N6 + H20.
Bi203 + 2C12 + 4KOH = Bi205 + 4KC1 + 2H20.

XLV THE METALS OF THE NITROGEN GROUP 943

The higher oxides are reduced when warmed with concentrated
hydrochloric or sulphuric acids :

Bi2O4 + 8HC1 = 2BiCl3 + 4H2O + C12
Bi205 + 3H2S04 - Bi2(S04)3 -f 3H20 + O2.

On fusing bismuth trioxide with caustic potash in air, a brown
mass of potassium bismuthate, KBi03, is formed. This is hydrolysed
by water, and Bi2O5 is precipitated. The higher oxides of bismuth
therefore show acidic properties. Potassium bismuthate is used
as an oxidising agent.

Bismuth salts. — The most important salts, the nitrate, Bi(N03)3,
and the basic nitrate, Bi(OH)2NOs, have already been described.

Bismuth sulphate, Bi2(SO4)3, is obtained by dissolving the metal
in hot concentrated sulphuric acid. It forms a basic sulphate,
Bi(OH)4S04, sparingly soluble, on addition of water. On heating,
this forms yellow (BiO)2S04, bismulhyl sulphate. A double salt,
KBi(S04)2, is formed with potassium sulphate. If sodium thio-
sulphate is added to a solution of a bismuth salt a clear solution
containing sodium bismuth thiosulphate, Na3Bi(S203)3, is formed.
This does not react with iodine. On adding a potassium salt and
alcohol to the solution, a sparingly soluble yellow precipitate of the
potassium salt, 2K3Bi(S203)3,H20, is formed, and the reaction may
be used for the detection of potassium. The solution of the sodium
salt quickly decomposes and deposits a black precipitate of bismuth
sulphide, Bi2S3.

The basic carbonate, 2(BiO)2C03,H20, is prepared by precipitating
a solution of the nitrate with ammonium carbonate ; on drying at
100° it loses water. Bismuth phosphate, BiP04, and pyrophosphate,
Bi4(P207)3, are obtained by precipitation. The latter fuses to a
glassy metaphosphate on heating. Bismuth does not readily combine
with arsenic or phosphorus.

Bismuth sulphide, Bi2S3, is obtained in crystals by fusing bismuth
with sulphur, or as a brownish-black precipitate when sulphuretted
hydrogen is passed into a solution of a bismuth salt. The pre-
cipitate dissolves in nitric acid, and in boiling concentrated hydro-
chloric acid, but not in alkalies or yellow ammonium sulphide,
since it does not, like the sulphides of arsenic, antimony, and tin,
form thio-salts in this way. The latter are produced by dissolving
the sulphide in concentrated sodium sulphide, or by fusion with
sulphides. The salts KBiS2 and NaBiS2 form fine crystals with a
metallic lustre, rapidly oxidised in the air. On diluting the solution
in sodium sulphide, Bi2S3, is reprecipitated. Precipitated Bi2S3
dissolves in water to the extent of 0-2 mgm. per litre.

Halogen compounds of bismuth.— Bismuth trichloride, BiCl3, is
formed as a soft, white, crystalline substance, m.-pt. 227°, b.-pt.
428°, on passing an excess of chlorine over bismuth. Its vapour

944 INORGANIC CHEMISTRY <HAI>.

density corresponds with the formula BiCl3. The trichloride is
also formed by dissolving bismuth in aqua regia. On cooling,
crystals of BiCl3,H2O are deposited. The solution in concentrated
hydrochloric acid deposits crystals of chlorobismuthous acid, H2BiCl5 ;
at 0° 2BiCl3,HCl,3H2O is deposited, stable at the ordinary tem-
perature. Salts of H2BiCl5, as well as of HBiCl4 and HBi2Cl7, have
been prepared.

A solution of bismuth chloride when poured into water gives a
white precipitate of bismuth oxychloride, or bismuthyl chloride,
BiOCl. This is deposited when any bismuth salt is added to a
solution of sodium chloride ; it resembles silver chloride in becoming
grey on exposure to light.

On heating BiCl3 with excess of bismuth, or by heating bismuth with
calomel at 250°, a black dichloride, BiCl2, corresponding with Bi,O2, is
formed. It is decomposed by water :

3BiCl2 + 2H20 = 2BiOCl + Bi + 4HC1.

The tribromide, BiBr3, is formed from the elements in golden-yellow
crystals, decomposed by water into white BiOBr. Bismuth tri-iodide,
BiI3, is a black powder obtained by adding bismuth oxide to a solution
of iodine in stannous chloride saturated with HC1. It is slowly decom
posed by water, forming red BiOI. Bismuth iodide dissolves i
hydriodic acid, forming iodobismuthous acid, HBiI4,4H2O, and
alkali iodides, forming red crystalline salts, e.g., KBiI4. Bismut
fluoride, BiF3, is a white powder obtained by evaporating a solution o
Bi2O3 in HF. With excess of oxide, BiOF is formed.

The atomic weight of bismuth, 2064 (H = 1) has been determinec
by various methods : conversion of Bi into Bi2O3, reduction o
Bi203 to Bi, conversion of BiBr3 into AgBr, and the conversion
Bi into Bi2(SO4)3.

Bismuth hydride. — By the action of concentrated hydrochloric acid
on an alloy of equal parts of bismuth and magnesium, hydrogen is
obtained which on passing through a heated tube deposits a brown
mirror of bismuth in front of the heated spot, and a fainter one behind,
indicating that traces of the gaseous bismuth hydride (?BiH3) are
formed. Thorium C, an isotope of bismuth, when deposited on mag-
nesium, also gives a radioactive gaseous hydride.

The rare metals of Group V. — Vanadium, niobium (or columbium),
and tantalum form acidic oxides of the general type R2O6, and corre-
sponding salts, usually meta-salts, MRO3 ; . e.g., ammonium meta-
vanadate, NH4VO3. Vanadium iorms a complete series of oxides,
V2O8, V2O4, V2O3, VO, and V2O, analogous to the oxides of nitrogen.
Compounds of these are produced by reducing a solution of V2O5 in
dilute sulphuric acid with sulphur dioxide (blue, V2O4), magnesium

XLV THE METALS OF THE NITROGEN GROUP 945

(green, V2O3), and zinc (lavender, V2O). The metal is obtained from
V2O5 and carbon in the electric furnace ; it is added to special steels.
The chlorides, VC14, VC13, and VC12, and an oxychloride,VOCl3 (of. POC13),
are known.

Niobium and tantalum are very rare elements : they form double
fluorides, K2TaF7, and 2KF,NbOF3,H2O. Metallic tantalum, obtained
by heating the oxide Ta2O5 with aluminium in a vacuum electric fur-
nace, is very resistant to acids, and has a high melting point (2850°) :
it was formerly used for electric lamp filaments.

EXERCISES ON CHAPTER XLV

1. How are antimony and bismuth obtained ? For what purposes
are the elements, and their compounds, used ?

2. Describe the preparation of antimony hydride. How may it be
distinguished from arsenic hydride ? What is known of the hydrides >
of the other elements of this group ?

3. Starting with antimony, how would you prepare : (a) the chlorides,
(6) the oxides, (c) the sulphides ? Compare the properties of these
compounds with the corresponding compounds of phosphorus and
arsenic.

4. Describe the preparation of four typical compounds from metallic
bismuth, mentioning their chief properties and uses.

5. Contrast the properties of the oxides of nitrogen, phosphorus,
arsenic, antimony, and bismuth, with special reference to their acidic
and basic character, and stability. • *•>

6. Describe the preparation and properties of the sulphides of anti-
mony and bismuth. What is the action of ammonium and sodium
sulphides on them ?

7. How are the chlorides of antimony and bismuth prepared ? What
is the action of water on these compounds ?

3 P

CHAPTER XLVI

THE METALS OF THE SULPHUR GROUP

The metals of Group VI. — Group VI of the Periodic System com-
prises eight elements :

Odd series. Even series.

Oxygen, O = 15-87 Chromium, Cr =51-6

Sulphur, S == 31-81 Molybdenum, Mo = 95-2

Selenium, Se=78-6 Tungsten, W =182-5

Tellurium, Te = 126-5 Uranium, U =236-3

At first sight no obvious resemblances exist between the elements
of the odd and even series. The former are all non-metals ; the
latter all metals. If we take sulphur as representative of the odd
series, and chromium as typical of the even series, however, a closer
examination of their chemical properties reveals many points of
similarity. Both form acidic oxides, R03, the salts of which are
isomorphous, and have similar formulae :

S03 K2S04 (K20,S03) K2S207 (K20,2S03)
Cr03 K2Cr04 (K20,Cr03) K2Cr207 (K20,2Cr03)

Polysulphates and polychromates also exist, containing more than
2R03 to one molecule of basic oxide.

Both elements form stable oxy chlorides, R02C12, hydrolysed by
water, but there is no chloride of chromium corresponding with
S2C12. The stable chloride of chromium is CrCl3, corresponding with
FeCl3 and A1C13, and chromium shows many resemblances to
aluminium and iron. The metals chromium and iron are similar,
and the hydroxides, A1(OH)3, Cr(OH)3, and Fe(OH)3, are all pre-
cipitated in a gelatinous form by adding ammonia to solutions of
the salts. Chromium hydroxide, however, appears to have the
formula Cr20(OH)4. These three metals are classed together in the
same group in qualitative analysis. The analogy betwreen iron and
chromium is also seen in the formation of ferrates, e.g., K2Fe04
(red), and chromates, e.g., K2Cr04 (yellow). The compounds
CrX2 are also closely analogous to the ferrous salts, and differ
from the corresponding sulphur compounds.

The elements molybdenum and tungsten resemble chromium in

CH. XL vi " THE METALS OF THE SULPHUR GROUP 047

their chemical properties : uranium differs somewhat from its com-
panions, since its stable salts are derived from a radical UO2, uranyl.
Molybdenum and tungsten form a number of complex acids with
phosphoric acid, etc.

( CHROMIUM . Cr — 5 1 -0.

Chromium. — A red Siberian mineral containing lead was described,
under the name of crocoisite, by J. G. Lehmann in 1762, but its com-
position was only elucidated in 1797 by Vauquelin and by Klaproth,
who found that it was a lead salt of chromic acid, Cr03) viz.,
PbO,Cr03, or PbCrO4. The name chromium (Greek chroma — colour)
was given to the element because it forms a large number of coloured
compounds. Metallic chromium was obtained in an impure state
by Vauquelin by reducing the sesquioxide Cr203 with carbon at a
white heat.

The commonest ore of chromium is chromite, or chrome-ironstone,
which is ferrous chromite, FeO2O4, or FeO,Cr203, a spinel (p. 891).
Rarer minerals are chrome-ochre, Cr2O3, and chromitite, Fe203,O203.

The chromates derived from the acidic trioxide, CrO3, are yellow or
red ; the chromic salts, derived from O203, are violet or green ; the
chromous salts, derived from CrO, are usually blue. The colours are
more intense in the hydrated salts.

Chromite is imported mainly from Asia Minor, Rhodesia, and New
Caledonia. It occurs in masses with a granular fracture, is very
refractory, and is made into chrome bricks used for furnace linings,
or to separate the silica bricks outside from the magnesia bricks
inside the basic hearth steel furnace (p. 981). Chromite is the source
of chromium compounds. If reduced with carbon in the electric
furnace, ferrochrome, iron with 60 per cent, of chromium, is formed,
which is used in the manufacture of chrome-steel, This contains 60
parts Cr, 36Fe, and 4 molybdenum, and is free from carbon. It is
not attacked by acids, and is usually known as " rustless steel." An
alloy of chromium, nickel, and iron is used for making armour-plates.

When powdered chromite is heated to bright redness with lime
and a little soda in contact with air, calcium chromate, CaCr04, is
formed. The residue is treated with hot sodium carbonate solution,
and the filtrate, containing sodium chromate, Na2Cr04, evaporated.
Sulphuric acid is then added, when sodium sulphate is precipitated
and a deep red solution of sodium dichromate, Na2Cr207, obtained,
from which deliquescent crystals of Na2Cr207,2H2O are deposited
after evaporation. In an older process, a mixture of chromite,
lime, and potassium carbonate was heated : 4FeCr2O4 + 8K2C08 +
7O2 = 2Fe203 + 8K2CrO4 + 8C02. The yellow solution of potass-
ium chromate, K2CrO4, obtained on addition of water to the mass
was treated with sulphuric acid, and potassium dichromate, K2Cr0O 7,

3 P 2

948

IXORGASTIC CHEMISTRY

CHAP.

obtained readily in bright red crystals : 2K2CrO4 4- H2S04 =
K2Cr2O7 4- K2SO4 j H2O. The sodium salt' is much cheaper and
more soluble, but may be converted into potassium dichromate by
treatment with potassium chloride. Chromates may also be
obtained by electrolysing alkalies with an anode of ferrochrome and a
cathode of porous copper oxide. Chromates and dichromates are
used as oxidising agents, as mordants in dyeing, and in preparing
insoluble pigments, e.g., lead chromate, PbO04.

EXPT. 329. — Fuse a little powdered chromite with sodium peroxide
in a nickel crucible. Extract the cooled mass with water. A yellow
solution of sodium chromate is obtained. This is converted into a red
solution of the dichromate when sulphuric acid is added.

Metallic chromium. — Chromium is obtained by reducing chromium
sesquioxide with aluminium in the thermit process (p. 894) : O203
4- 2A1 = 2A12O3 4- Cr. The reaction evolves so much heat that the
alumina fuses, and on cooling forms crystalline corubin. The
chromium forms a fused mass below the alumina,
and has a purity of 99-5 per cent. Cr. It contains
a little iron and silicon.

In Goldschmidt's thermit process (1898) a
mixture of the oxide and aluminium powder in a
crucible is ignited by a small cartridge of barium
peroxide and magnesium powder placed in a
depression in the mixture. This is kindled by a
small piece of magnesium ribbon.

EXPT. 330. — A tin canister, 10 in. by 6 in., is
filled with coarsely-powdered fluorspar, and a
depression 2 in. X 8 in. made in it by a large
test -tube. The mixture of dry chromic oxide and aluminium powder
is pressed into this, and the (BaO2 4 Mg) igniter placed on the top
(Fig. 418). The fluorspar is a good heat insulator, so that a fused mass
is obtained even with small amounts of material. A mixture of
aluminium powder with an equal, or double, weight of calcium turnings,
corresponding with the oxygen of the oxide, acts even more effectively
than aluminium alone.

When chromium oxide is reduced by carbon at very high tempera-
tures, the carbides, Cr5C2 and Cr3C2, are formed.

Pure chromium is obtained by electrolysing a solution of chromic
chloride, CrCl3, with a mercury cathode, and heating the amalgam
in a vacuum to remove mercury.

Chromium is a silver- white, hard, crystalline metal, sp. gr. 6-92,
m.-pt. 1615°, b.-pt. 2200°. It burns brilliantly in the oxy-hydrogen
flame, forming the sesquioxide, Cr203. Chromium dissolves in
dilute sulphuric and hydrochloric acids, especially on heating, form-

Fia. 418.— Arrange-
ment for Thermit
Reaction.

XLVI THE METALS OF THE SULPHUR GROUP 949

ing blue solutions of chromous salts: Cr -f 2HC1 — H2 -f CrCl2. The
blue solutions formed rapidly absorb oxygen on exposure to air,
forming green solutions of chromic salts: 4CrCl2 -f 4HC1 -f 02 —
4CrCl3 -f 2H20. Dilute nitric acid also dissolves chromium, but in
the concentrated acid it becomes passive, and is then unattacked by
dilute acids. Passivity is also induced by exposure to air, or dipping
in chromic acid. It is destroyed by touching the metal under the
surface of dilute sulphuric acid with zinc. A film of oxide may be
the cause of passivity (p. 985). Chromium decomposes steam at a
red heat : 2Cr + 3H20 '= Cr203 + 3H2. The finely-divided chro-
mium left on heating the amalgam is pyrophoric ; it combines with
nitrogen on heating, forming the nitride ON.

Chromous salts. — The chromous salts, CrX2, contain bivalent
chromium, and yield the ion Cr". They are powerful reducing
agents. Chromous salts are formed by dissolving the metal in acids,
or by reducing chromic salts with zinc and
dilute acid : Cr'v -f H = Cr' * + H\

EXPT. 331.— Place 50 gm. of granulated
zinc and 50 gni. of finely-powdered potass-
ium dichromate in a flask of 3 litres capacity
fitted with a tap -funnel, and a wide delivery
tube dipping under water (Fig. 419). Add
through the funnel a mixture of 300 c.c. of
concentrated hydrochloric acid and 200 c.c.
of water. A violent reaction occurs, the
liquid first becoming green (CrCl3) and then
blue (CrCl2). A saturated solution -of sodium
acetate is then added (containing 92 gm. of
sodium acetate crystals), when a red precipi- FIG. 419.— Preparation of
tate of chromous acetate, Cr(CH3-CO2)2, is

thrown down. This is fairly stable ; it is filtered off, rapidly washed
with water saturated with carbon dioxide, and transferred to the flask.
The air is expelled from the latter by hydrogen, and the solid dissolved
in hydrochloric acid. A blue solution of chromous chloride is formed.
This is cooled in ice. and a current of hydrogen chloride gas passed
through. Chromous chloride, CrCI2,4H2O, is precipitated in blue
needles.

Anhydrous chromous chloride is obtained by heating chromic
chloride in hydrogen : 2CrCl3 + H2 = 2CrCl2 -f 2HC1, or metallic
chromium hi hydrogen chloride. It forms white, silky needles. The
vapour density at 1300° is 113 (CrCl2 = 63-5, Cr2Cl4 == 127) ; at
1600° it is 89 : Cr2Cl4 =r 2CrCl2.

Chromous sulphate, CrS04,7H2O, is obtained in fine blue crystals
isomorphous with ferrous sulphate by dissolving the acetate in dilute
sulphuric acid. It forms double salts, e.g., K2S04,CrSO4,6H2O.

950 INORGANIC CHEMISTRY CHAP.

The ammoniacal solution of CrS04 absorbs acetylene and nitric
oxide.

Caustic soda added to a solution of a chromous salt gives a brownish-
yellow precipitate of chromous hydroxide, Cr(OH)2, which is readily
oxidised in air and in the moist state evolves hydrogen ; in both
cases chromic hydroxide is formed : 2Cr(OH)2 -f *2H2O = 2Cr(OH)3
-j- H2. Chromous oxide, CrO. cannot therefore be obtained by heat-
ing the hydroxide ; it is formed as a black powder on exposure of
chromium amalgam to air. Chromous carbonate, CrC03, is formed
as a yellow precipitate when sodium carbonate solution is added to
a solution of chromous chloride.

Chromic oxide, Cr203. — The chromic salts are stable compounds
containing tervalent chromium, and correspond with the very stable
basic chromium sesquioxide, or chromic oxide, O203. Chromic oxide is
produced as a green powder by heating chromic hydroxide :
2Cr(OH)3 = Cr203 + 3H20, ammonium dichromate : (NH4)2Cr207
— Cr2O3 -j- N2 + 4H20, or potassium dichromate with sulphur :
K2Cr2O7 -f S = K2S04 -f- Cr203. A very fine green oxide is pro-
duced by gently heating mercurous chromate : 4Hg2CrO4 =
8Hg -J- 2Cr2O3 -}- 502. The oxide is obtained in dark-green, hard,
hexagonal crystals by fusing the amorphous oxide with calcium
carbonate and boron trioxide, by igniting a mixture of potassium
dichromate and common salt, or by passing the vapour of chromyl
chloride, Cr02Cl2 (q.v.}, through a red-hot tube.

The oxide produced by gentle ignition of the hydroxide, or of ammon-
ium dichromate, is soluble in acids, and acts as a powerful catalytic
agent (e.g., in the oxidation of ammonia, p. 575). The crystalline form,
or the strongly- ignited oxide, is insoluble in acids, and inactive. It
may be brought into solution by fusing with potassium hydrogen
sulphate.

Chromic oxide is very refractory (m.-pt. 1990°), but dissolves in
fused borax or glass, giving to it a green colour, which becomes blue
if strontium is present ; this is applied in tinting glass and painting
porcelain . The oxide is also used as a permanent green oil paint
under the name of chrome-green.

Chromic hydroxide is formed by precipitating a solution of a
chromic salt with caustic potash, soda, or ammonia. As ordinarily
prepared by precipitating a hot solution with alkali, it is a green,
flocculent precipitate, which appears to have the composition
Cr2O(OH)4, i.e., Cr203,2H20. By precipitating a cold solution of a
violet chromic salt (q.v.} with ammonia, a pale blue precipitate,
which yields Cr(OH)3.2H20 when dried over sulphuric acid, is
formed. When heated in hydrogen at 200°, this gives CrO (OH) ; at a
red heat this passes into insoluble Cr203 with incandescence. The
blue hydroxide dissolves in caustic soda, giving a grass-green solution

XLVI THE METALS OF THE SULPHUR GROUP 951

which may contain a soluble chromite, Na2CraO4, or Na20,O203.
Natural chrome-ironstone is ferrous chromite. FeCr2O4. The green
solution may. however, be merely a colloidal solution of the hydr-
oxide, since^all the chromium hydroxide is deposited on boiling. A
dark green colloidal solution is obtained by dialysing a solution of
the freshly -precipitated hydroxide in chromic chloride solution. It
can be boiled, but is coagulated by salts.

By fusing together equimolecular amounts of potassium dichromate
and crystallised boric acid and lixiviating with water, a brilliant
green powder, used as a pigment under the name of Guignet's green,
is left. This is usually supposed to be the hydroxide Cr2O(OH)4, but
always contains boric acid (3O2O3,B2O3,4H2O).

Chromic chloride, CrCl3. — Anhydrous chromic chloride is obtained
as a sublimate of scaly, peach-blossom coloured crystals when
chlorine is passed over a mixture of chromium sesquioxide and carbon
heated to whiteness : Cr2O3+3C+3Cl2 = 2CrCl3H-3CO. The crystals
volatilise at 1065°, giving a density corresponding with CrCl3. They
are almost insoluble in cold water, but readily dissolve in presence of
a trace of chromous chloride, giving a green solution.

Three crystalline hydrates, CrCl3,6H20, are known, two green and
one violet. Precipitated chromium hydroxide dissolves in concen-
trated hydrochloric acid to form a dark green solution of chromic
chloride. If this is cooled in ice and saturated with hydrogen
chloride, small emerald-green crystals separate. When these are
dissolved in their own. weight of water, warmed to 80°, and then
cooled to 0°, the solution deposits greyish-blue crystals which dissolve
in cold water to give a violet solution. If the crystals are not
filtered off, and hydrogen chloride in excess is passed into the solution,
a second green form is precipitated.

In solution, the greyish -blue form gives three chloride ions, since
all the chlorine can be precipitated with silver nitrate. The first
green form gives only two chloride ions, and readily loses a molecule
of water. The second green form gives only one chloride ion, and
readily loses two molecules of water. Werner (p. 1010) represents
the constitution of the three forms as follows :

greyish-blue : [Cr(OH2)6]Cl3 ;

first green : [Cr(OH2)5Cl]Cl2 + H20 ;

second green : [Cr(OH2)4Cl2]Cl + 2H2O. -

The atoms or molecules inside the square brackets are directly
combined with the metal atom, and are not ionisable, whereas those
outside are ionisable, or readily split off. The number of atoms or
molecules associated with the metal atom is always six.

Chromic fluoride, CrF3, is obtained in needles by passing HF over
CrCl3. The hydrated form, OF3,9H2O, is precipitated on adding

952 INORGANIC CHEMISTRY CHAP.

NH4F to a solution of Cr2(SO4)3. It forms a violet solution with
hydrochloric acid. The bromide, CrBr3, and two hydrates, CrBr3,6H2O,
are formed similarly to the chloride. The iodide is unknown.

Chromic sulphate. Cr2(S04)3. — This salt is obtained in violet
crystals by allowing a mixture of equal parts of concentrated sul-
phuric acid and chromic hydroxide (dried at 100°) to stand for some
weeks in a loosely-stoppered bottle. If its solution is precipitated
with a little alcohol, violet octahedra. Cr2(S04)3.18H20, are deposited.
With excess of alcohol the anhydrous sulphate is thrown down.
Chromic sulphate combines with sulphates of the alkali-metals,
forming chrome alums.

Potassium chromium sulphate, ordinary chrome alum, has the
formula K2S04,Cr2(S04)3,24H20. It is obtained by reducing a
solution of potassium dichromate acidified with sulphuric acid, and
hence often separates in purple octahedral crystals on the carbon
poles of bichromate cells after use. In these cells zinc and carbon
plates are immersed in a solution of potassium dichromate in dilute
sulphuric acid. The hydrogen liberated on the carbon plates is
oxidised by the chromic acid.

EXPT. 332. — Dissolve 20 gm. of potassium dichromate in 100 c.c. of
hot water, and after cooling add carefully 37 c.c. of concentrated sul-
phuric acid. Pass sulphur dioxide through the solution until the red
colour, which at first changes to brown and then to olive -green,
has become pure dark green :

K2Cr207 + H2S04 + 3S02 = K2SO4 + Cr2(SO4)3 + H2O.
Evaporate the green solution to about one -fourth its volume, set aside
in a covered dish for some time, and observe the formation of purple
octahedral crystals of chrome alum. Instead of sulphur dioxide, alcohol
may be used in the reduction : the alcohol is oxidised to aldehyde,
C2H4O.

Chrome alum is formed as a by-product in the oxidation of
anthracene, C14H10, to anthraquinone, C14H802, by sulphuric acid
and potassium dickromate. It is used in dyeing and calico-printing,
and in tanning.

In chrome-tanning the hides are steeped in a solution of chrome
alum. Chromic hydroxide is absorbed by the gelatine of the hide,
forming a green, insoluble substance. The dry leather may then be
treated with melted paraffin wax to render it waterproof (" driped ").

A solution of chrome alum in cold water has a dull bluish -red
colour ; on heating to 70° it becomes green. Barium chloride
precipitates the sulphate in the violet solution completely, whilst
the green solution is not completely precipitated. If the green
solution is allowed to stand for some time in the cold, it becomes
violet again and barium chloride precipitates all the sulphate.

XL vi THE METALS OF THE SULPHUR GROUP 953

A green variety of chromic sulphate is formed by heating the
violet crystals, Cr2(S04)3J8H20, at 90° until they have the composi-
tion O2(S04)3,8H20. The solution is not precipitated either by
alkalies or barium chloride. By the action of sulphur dioxide
on chromic acid below 0°. Colson has obtained a green salt,
Cr2(SO'4)3;6H20, the freshly-prepared solution of which does not
react with barium chloride. On standing, the solution is trans-
formed successively into green substances from which barium
chloride precipitates one-third and two-thirds of the sulphate, and
finally into a violet solution, completely precipitated by barium
chloride. Werner represents the four forms as follows :

[Cr^SO^EUOy + 3H20 ; [Cr2(S04)2(H20)4]S04 + 2H20 ;
[Cr2(S04)(H20)5](S04)2 + H20 ; and [Cr2(H20)6)](S04)3.

A number of complex chromic -sulphuric acids and other salts are
known.

Chromium nitrate, Cr(NO3)3,9H2O is formed from the hydroxide and
nitric acid. The phosphate, CrPO4, is formed by precipitation of
chromium salts with sodium hydrogen phosphate as an amorphous
violet precipitate. On standing for a day or two in contact with the
solution this is converted into a violet crystalline hexahydrate,
CrPO4,6H2O. If allowed to stand for a week in the solution, the
amorphous precipitate is converted into a green amorphous tetra-
hydrate, OPO4,4H2O A green crystalline tetrahydrate is formed
by boiling the violet hexahydrate with water for half an hour. On
heating, all the hydrates give a black powder of CrPO4. Chromic
acetate is obtained as a green solution, used as a mordant, by dissolving
the hydroxide in acetic acid.

Chromium sulphide, Cr2S3, is obtained by heating sulphur with
chromium, or CrCl3 in H2S. By adding a solution of ammonium
sulphide to a chromic salt, the hydroxide is precipitated, as the
sulphide is completely hydrolysed by water : 2CrCl3 -f 6H2O +
3(NH4)2S = 2Cr(OH)8 + 6NH4C1 + 3H2S.

Chromium trioxide or chromic acid, Cr03.— By the action of con-
centrated sulphuric acid on a solution of a dichromate, red chromium
trioxide, Cr03, is obtained. This substance is often called " chromic
acid," although this should have the formula H2CrO4. True
chromic acid is said to be formed by warming the trioxide with a little
water and cooling, but if it exists it is very unstable. The aqueous
solution of chromium trioxide has a red colour and is strongly acid.
The depression of freezing point and the conductivity show that the
solution contains dichromic acid, H2O207, which is not known in the
pure state.

EXPT. 333. — Dissolve 50 gm. of K2O2O7 in 85 c.c. of water, and to
the cooled solution add slowly 70 c.c. of concentrated H2SO4. Allow

954 INORGANIC CHEMISTRY CHAP.

to stand for twelve hours and pour the liquid off the crystals of acid
potassium sulphate which have separated : K2O2O7 -f 2H2SO4 =
2CrO3 + 2KHSO4 -f H2O. Heat to 85°, add 25 c.c. of sulphuric
acid and sufficient water just to dissolve the CrO3 separating. Allow
to stand twelve hours, and decant the liquid from the crystals of CrO3.
\Vash the latter in a Biichner funnel containing asbestos with pure
nitric acid, and heat to 60-80° in a current of air in a tube to remove
the adhering nitric acid.

Chromium trioxide forms a deliquescent red woolly mass or red
lustrous rhombic prisms. It melts at 393° to a dark red liquid,
solidifjdng on cooling to a reddish-black mass with a metallic lustre.
At 250° it loses oxygen : 4Cr03 = 2Cr203 -f 302 ; a little of the
trioxide sublimes. Chromium trioxide is a very powerful oxidising
agent. Alcohol dropped on it catches fire ; the concentrated
solution is reduced by sugar, oxalic acid, paper, cork, etc. It
oxidises sulphur dioxide, hydrogen sulphide, stannous chloride,
arsenious oxide, ferrous salts, etc. In acid solutions the reduction
always proceeds to the stage of a chromic salt : 2Cr03 = Cr203 +
30. A solution of potassium dichromate mixed with sulphuric acid
is very often used in organic chemistry as an oxidising agent ; a
solution of chromium trioxide in glacial acetic acid (which is not
oxidised) is also applied.

Chromates. — Chromic acid in its salts shows the closest analogies
to sulphuric acid, and its formula may be written Cr02(OH)2. It
forms normal chromates (e.g., K2Cr04), and dichromates (e.g.,
K2Cr207), analogous to sulphates and disulphates. Acid chromates,
e.g., KHCrO4, are not known, but by the action of excess of Cr03. or
by boiling the dichromate with nitric acid, trichromates (e.g.,
K2Cr3Oto, or K20,3Cr03) and tetrachromates (e.g., K2Cr4013, or
K20,4Cr03) are formed as red crystals.

Normal potassium chromate, K2CrO4, is obtained in lemon-yellow
crystals by neutralising a solution of chromic acid or the dichromate
with caustic potash or potassium carbonate, and evaporating. It is
isomorphous with potassium sulphate. Potassium dichromate, K2Cr207
(p. 947), may be obtained by adding the requisite amount of sul-
phuric acid to a saturated solution of the normal chromate, and
crystallises out on cooling in garnet -red crystals. The solubilities
of the two salts are as follows, in 100 parts of water : —

0° 30° 60° 105-8° 104-8°

K2CrO4 . . . . 54-57 65-13 74-60 88-8 (b-pt.)
K2Cr2O7 , . . . 4-64 18-13 45-44 108-2 (b.-pt.)

Both salts are non-deliquescent : they crystallise without water.
Sodium chromate, Na2CrO4,10H20, and dichromate, Na2Cr207,2H20,
made on a large scale, are deliquescent. A solution of sodium
chromate is produced by triturating moist chromium hydroxide

XLVI THE METALS OF THE SULPHUR GROUP 955

with sodium peroxide. Ammonium chromate, (NH4)2Cr04. is
unstable ; it is obtained by crystallising solutions containing excess
of ammonia. Ammonium dichromate, (NH4)2Cr207. is readily
obtained by adding ammonia to the requisite amount of chromium
trioxide in solution. It forms orange-red crystals which decom-
pose violently on heating, evolving nitrogen and steam and leaving
a voluminous dull-green mass of chromic oxide. All soluble chro-
mates are poisonous.

Metallic chromates, if soluble, are formed from the oxides and
chromic acid ; they are often insoluble and can then be prepared
by double decomposition. The most important sparingly soluble
chromates are :

Silver chromate : Ag2CrO4 ; brick -red, soluble in acids and ammonia.

Barium chromate : BaCrO4 ; yellow, insoluble in acetic acid, soluble
in mineral acids.

Zinc chromate (basic) : Zn2(OH)2CrO4,H2O ; yellow.

Lead chromate : PbCrO4 (chrome-yellow — used as a pigment) — pre-
cipitated from Pb(NO3)2 and K2O2O7 ; soluble in nitric acid and in
caustic potash.

Basic lead chfomate : Pb?CrO5 (chrome-red — used as a pigment) —
by digesting PbCrO4 with cold caustic soda ; mixed with PbCrO4 it
forms the pigment chrome-orange.

Bismuth chromate (basic) : 2(BiO)2CrO4,Bi2O3 ; lemon yellow.

Chromic chromate : Cr2O3,CrO3 = 3CrO2 (chromium dioxide) — by
heating chromic nitrate, or precipitating a chromic salt with a chromate.

Potassium dichromate in acid solution liberates iodine from
potassium iodide : K2Cr2O7 + 7H2SO4 + 6KI = O2(SO4)3 +
4K2S04 + 7H2O + 3I2. It is used in volumetric analysis for the
estimation of ferrous iron. In acting as an oxidising agent it is
reduced to a chromic salt : K20,O206 = K20 + Cr2O3 + 30. One
gm. molecule therefore contains 3 atoms, or 6 equivalents, of avail-
able oxygen ; a decinormal solution, containing 0-1 equivalent of
available oxygen per litre, is produced by dissolving 4-913 gm. of
K2Cr207 in a litre of water. This oxidises ferrous salts in acid solu-
tion according to the equation : 2FeO + O = Fe2O3, hence 1
equivalent of oxygen (10 litres of iV/10K2Cr207) oxidises two
equivalents of ferrous iron, or 56 gm. The titration of the ferrous
salt is complete when a drop of the liquid, brought in contact with a
drop of freshly -prepared potassium ferricyanide solution on a white
plate, no longer gives a blue colour (p. 248).

Chromyl chloride, Cr02Cl2. — Chromium, and the other metals of
the chromium group, form oxychlorides, containing the bivalent
radicals R02, viz., RO2C12 :

chromyl chloride, CrO2Cl2 tungstyl chloride, WO2C12

molybdyl chloride, Mo02Cl2 uranyl chloride, U02C12

956 INORGANIC CHEMISTRY CHAP.

When a mixture of sodium chloride and potassium diohroinate is
distilled in a retort with concentrated sulphuric acid, a deep red
vapour is produced, which condenses to a nearly black liquid like
bromine. This is chromyl chloride, Cr02Cl2. If chromium trioxide is
dissolved in concentrated hydrochloric acid, and concentrated sul-
phuric acid added in small quantities at a time to the cooled liquid,
chromyl chloride separates, and may then be distilled : O03+2HC1 ^±
Cr02Cl2 + H20. It boils at 115°9°, and is decomposed violently by
water, "with production of chromic and hydrochloric acids. The
vapour density corresponds with the formula Cr02Cl2. Chromyl
chloride is a powerful oxidising agent, exploding in contact with
phosphorus (cf. Br2) and inflaming sulphur, ammonia, alcohol, and
many organic substances. Bromides and iodides do not produce
corresponding compounds when distilled with dichromate and sul-
phuric acid, but the free halogen is liberated : this may be utilised
in the detection of chlorides in presence of bromides and iodides,
since if the former is present the distillate, when collected in water,
produces chromic acid, and gives with lead salts a vellow precipitate
of PbCr04.

Chlorochromates. — When powdered potassium » dichromate is
dissolved in warm concentrated hydrochloric acid, and the liquid
cooled, or if chromyl chloride is added to a saturated solution of
potassium chloride, red crystals of potassium chlorochromate,
KCr03Cl, are formed :

K2Cr2O~ + 2HC1 = 2KCr03Cl + H20
CrO2Cl2 + KC1 + H2O = KCr03Cl + 2HC1.

This salt is known, after its discoverer, as Peligot's salt ; it probably

/Cl
has the constitution Cr02<f and is the salt of an unknown

chlorochiomic acid, [cf. chlorosulphonic acid, SO2(OH)C1] :

Cr02(OH)2 O02(OH)C1 CrO2Cl2.

(hypothetical) chromic acid, (hypothetical) ehloroehromie acid, ehromyl chloride.
The salt is partly decomposed by water, and decomposes
at 100° with evolution of chlorine (cf. p. 222) 4Cr02(OK)Cl =
K2Cr2O7 + O203 + 2KC1 + C12 + O2. When it is suspended in
ether and treated with dry ammonia, stable red crystals of the
amino-chromate, Cr02(OK)NH2, are formed. If this substance is
treated in an ethereal solution of ammonia with chlorine, brown
chromylamine, Cr02(NH2)2, is deposited.

Perehromic acid. — If hydrogen peroxide is added to an aqueous
solution of chromic acid, or of a chromate acidified with sulphuric
acid, a dark coloured liquid is produced which on agitation with
ether gives a deep indigo-blue colour to the latter (p. 340). This
blue ethereal solution contains a higher oxygen compound of chrom-

XLVI THE METALS OF THE SULPHUR GROUP 957

ium, called perchromic acid. On evaporation, or addition of alkalies,
oxygen is evolved, and a chromic salt is formed.

By" the action of organic bases (aniline, pyridine, etc.) on the blue
ethereal solution, deep-blue salts are formed which are explosive. These
have been represented as CrO4(OR),H2O2, i.e., derived from HCrO5,
or as acid salts, RH2OO7. derived from H3CrO7. From alkaline
chromate solutions and H2O2> red salts are obtained, of the formula
R3CrO8, which on treatment with acids evolve oxygen and form the
blue salts. Free perchromic acid is obtained by adding 97 per cent.
H2O2 to a solution of Cr03 in methyl ether cooled to — 30°, pouring off
the blue liquid from excess of CrO3, and evaporating in a vacuum at — 30°.
The dark blue, crystalline mass decomposes at a temperature slightly
above — 30°. Its composition corresponds with the formula
H3CrO8,2H2O, but the water may be constitutional and the formula
(OH)4Cr(OOH)3. The red salts may be anhydro-salts of the blue

°V
acid, viz., ^Cr(O-OH)3.

Molybdenum, Mo = 95-2. — The mineral molybdenite resembles
graphite in appearance, but was shown by Scheele (1778) to consist
of molybdenum sulphide, MoS2. When roasted in air it leaves a
residue of molybdenum trioxide, Mo03 (" molybdic acid "), which
dissolves in ammonia to form ammonium molybdate, (NH4)2Mo04.
The crystals obtained by evaporation (ordinary " ammonium molyb-
date ") are more complex, (NH4)6Mo7024,7H2O. Molybdenum and
tungsten show marked tendencies to form such complex compounds.
A solution of ammonium molybdate in nitric acid gives with phos-
phoric acid a canary-yellow precipitate of phosphomolybdic acid,
which when dried at 100° has the composition (NH4)3P04,12Mo03.
The chlorides MoCl5, MoCl4, MoCl3, and MoCl2 are known, as well aw
the hexafluoride, MoF6 (cf. SF6). Molybdenum is a white metal of
high melting point (2450°) obtained by reducing the trioxide with
aluminium (cf. Cr). Its alloy with iron (ferromolybdenum) is
prepared by reducing molybdenite with iron and carbon in the electric
furnace : steel containing 2 per cent, of molybdenum does not soften
on heating and is used for high-speed lathe tools.

Tungsten, W = 182-5. — The heavy mineral now called, scheelite
was found by Scheele in 1781 to be calcium tungstate, CaW04. A com-
moner mineral is wolfram, ferrous tungstate, FeWO4, found with tin-
stone in Cornwall (p. 10). If these minerals are boiled with hydro-
chloric acid, a yellow powder of tungsten trioxide, WO3 (" tungstic
acid "), remains. If this (or wolfram) is fused with sodium carbonate,
soluble sodium tungstate, a complex salt, Na10W12041,28H2O, is
obtained, which is used as a mordant and in rendering flannelette
non-inflammable. Colloidal tungstic acid is obtained by dialysing a

958 IXORGANIC CHEMISTRY CHAP.

.solution of sodium tungsijilr (o which hydrochloric acid has been
added. Phosphotungstic acid, obtained from sodium tungstate and
phosphoric acid, is soluble in ether and is used as a reagent for
alkaloids. Metallic tungsten is obtained by reducing the trioxide
with hydrogen at a red heat : ferrotungsten, obtained in the electric
furnace, is used for special steels (7-9 per cent, of W ; 2-3 per cent,
of Cr). Tungsten filaments (m.-pt. 3100°) are used in electric lamps.
The compounds WF6 (gas), WC16, WC15, WC14, and WC12 are
known.

Uranium, U — 236-3. — The black mineral pitchblende, found in
Bohemia, Saxony, East Africa, and Colorado, was found by Klaproth
(1789) to be the oxide of a metal which he called uranium : U308.
Other uranium minerals, e.g., carnotite. a vanadate of uranium and
potassium (23 per cent, of U) are found. All these ores contain traces
of radium (p. 1021). If pitchblende is dissolved in concentrated
sulphuric acid, the lead, etc., separated by H2S, and ammonia added
to the nitrate, a precipitate of ferric hydroxide and uranyl hydroxide,
U02(OH)2. is formed, from which ammonium carbonate dissolves
the uranium, forming a crystalline compound, U02C03,2(NH4)2C03,
which on ignition yields the pure oxide, U308. When this is dis-
solved in nitric acid, yellow, fluorescent crystals of uranyl nitrate,
U02(N03)2,6H20, commonly called " uranium nitrate," separate.
Uranium salts mostly contain the bivalent uranyl radical, U02.
They are used in photography and in making fluorescent glass. The
chlorides UC13, UC14, and UC13, and the fluoride, UF5, are known. The
oxychloride, U02C12 (cf. Cr02Cl2),is formed by heating the oxide with
charcoal in chlorine. The metal is obtained by reducing UC14 with
sodium. Alloys with iron are obtained in the electric furnace, and
used in making special steels.

EXERCISES ON CHAPTER XLVI

1. Discuss the inclusion of chromium in the sulphur group. With
what other elements does chromium show analogies ?

2. In what forms does chromium occur ? How is sodium dichr ornate
manufactured from chromite, and for what purposes is it used ?

4. How would you prepare (d) chromium trioxide, (b) chromyl
chloride, (c) chromium sesquioxide, (d) chrome alum, (e) chromoi
chloride, from potassium dichromate ? Describe the properties
these substances.

4. How are chromium, molybdenum, and tungsten obtained ? Fc
what purposes are they used ?

5. What are the general properties of (a) chromous salts, (b) chromic
salts ? In what forms does chromic sulphate exist ? What explana-
tion of the existence of these forms has been given ?

6. What happens when potassium dichromate is : (a) warmed with
concentrated hydrochloric acid ; (b) boiled with concentrated sulphuric

XLVI THE METALS OF THE SULPHUR GROUP 959

acid ; (c) treated, in a solution acidified with sulphuric acid, with sulphur
dioxide ; (d) treated with zinc and dilute hydrochloric acid ; (e) added,
in solution, to acidified hydrogen peroxide ?

7. How is perchromic acid obtained ? What is the formula of this
substance ?

8. How is ammonium molybdate obtained from molybdenite ? What
happens when this salt is added to a solution of a phosphate acidified
with nitric acid.

CHAPTER XLVH

MANGANESE. Mn = 5440

Manganese. — The position of manganese in the seventh group
of the periodic system is one of isolation. The only property in
which it shows analogies with the halogen elements is the formation
of a higher oxide. Mn207, which forms salts, permanganates, e.g.,
KMn04, isomorphous with perchlorates, e.g., KC104. The oxide
Mn2O7 is also volatile and explosive like C12O7. Both silver per-
chlorate and silver permanganate are sparingly soluble in water.

In its remaining compounds manganese shows much closer analo-
gies with chromium and iron, the two elements of adjoining groups
in the same series (see the periodic table). The metals are similar in
physical properties, and both manganese and chromium form basic
sesquioxides, dioxides, and acidic trioxides. Potassium chromate
(yellow), K2Cr04, and potassium manganate (green), K2Mn04, are
isomorphous. The salts corresponding with the sesqui oxide,
Mn2O3, e.g., Mn2(S04)3. are much less stable than those of chromium,
e.g., Cr2(S04)3. Manganese resembles iron in forming three oxides of
the types HO, R203, and R3O4, the first two of which form series of
salts. The manganous salts, however, are more stable than the
ferrous salts, e.g., they do not undergo oxidation on exposure to air.
Manganese resembles magnesium in forming a sparingly soluble
compound, MnNH4P04 (p. 859).

Manganese ores. — The most important ore of manganese is the
black dioxide, Mn02, known as pyrolusite. This is referred to by
Pliny as " magnesia." but was confused with an ore of iron, the
magnetic oxide, Fe304. The name pyrolusite (Greek pyr, fire :
luo, I dissolve) refers to the use of the mineral in decolorising glass.
The materials used in making glass usually contain iron, and the
ferrous silicate produced gives a green colour to the glass. If
pyrolusite is added hi small quantity, the ferrous silicate is oxidised
to ferric silicate, which has a pale yellow colour, neutralised by the
purple tinge imparted by the manganese. Pott in 1740 and Scheele
in 1774 investigated pyrolusite ; metallic manganese was first ob-
tained in an impure form by Gahn, by heating the mineral with
carbon : Mn02 + 2C = Mn -f 2CO.

960

CH. XLVII MANGANESE 961

Pyrolusite occurs in many localities such as Bohemia, Spain,
India, and North America. It is usually contaminated with barium,
often in the form of psilomelane, (Mn,Ba)0,2Mn02, and with ferric
oxide. Pyrolusite always contains less oxygen than corresponds
with the formula Mn02 ; if used for the manufacture of chlorine
by the Weldon process (p. 240). the ore is valued on its content of
available oxygen. Most of the ore is now used in smelting for ferro-
manganese, and the manganese content is of more importance.
About 700.000 tons of manganese ores are produced annually.

Less important manganese minerals are the oxides, braunite, Mn3O3 ;
hausmannite, Mn3O4 ; the hydrated sesquioxide, manganite, Mn2O3,H2O ;
hydrated dioxides, wad and psilomelane ; the carbonate, dialogite,
MnCO3 ; the silicate, rhodonite., MnSiO3 ; and the sulphide, alabandite,
MnS. The deposits of hydrated oxides are sedimentary (precipitates,
or derived from oxidation by plants, etc., in lakes), or metamorphic
(derived from the weathering of rocks).

Metallic manganese. — Impure manganese is obtained by reducing
the dioxide with carbon, as described above. If less than the
theoretical amount of carbon is used, and the mixture is heated in
the electric furnace, a purer metal (nearly free from carbon) is
produced : Mn02 + 20 = Mn -f 2CO. A purer metal is obtained
by reducing the pure dioxide with aluminium in the thermit process
(p. 948) : 3MnO2 + 4A1 = 2A1203 + 3Mn. The purest metal is
obtained by electrolysis of a concentrated solution of manganous
chloride with a mercury cathode, and distilling of the mercury in
a vacuum at 250°.

Manganese is a greyish-white, hard, and brittle metal, sp. gr.
7-4, easily oxidised by air unless it contains iron. It has a high
melting point (1245°), but volatilises readily in the electric furnace
above 2000°. The metal decomposes water even in the cold, with
evolution of hydrogen, and readily dissolves in dilute acids, forming
manganous salts : Mn -+- H2S04 = MnSO4 -f H2. It unites directly
with nitrogen above 1210°, forming a nitride, Mn5N3 (Mn3N2 is
formed by passing ammonia over the heated metal), and with carbon
in the electric furnace, forming a soft carbide, Mn3C.

Alloys of iron and manganese, obtained in the blast furnace, are
ferromanganese (30-80 per cent, of Mn) and spiegel (5-20 per cent, of
Mn); they are used in making manganese steel, which may contain up
to 13 per cent, of Mn, and is very hard and tough. It is used for
the jaws of rock-crushers, and for machinery. Manganese bronze is
copper alloyed with 1-2 per cent, of Mn, 8-15 per cent, of Sn,
and 0-5 per cent, of Zn. Alloys of copper and zinc with small
quantities of manganese resemble German silver. Manganin is
an alloy of 83 parts of Cu, 13 of Mn, and 4 of Ni. It is used for
resistance coils, since its electrical resistance is only slightly affected

962 INORGANIC CHEMISTRY CHAP.

by temperature after it has been heated repeatedly at 120°. An
alloy of 55 of Cu, 15 of Al, and 30 of Mn is magnetic.

Oxides of manganese. — Manganese forms six oxides, the lower
oxides being basic, and the higher acidic, which give rise to corres-
ponding series of salts : —

ii
Manganous oxide, MnO ; strongly basic, forming manganous salts,

MnS04. n m

Mangano-manganic oxide, Mn304 or MnO.Mn203 ; a mixed oxide.

in
Manganic oxide, Mn203 : feebly basic, forming manganic salts,

Mn2(S04)3. iv

Manganese dioxide, Mn02 ; feebly acidic, forming manganites,

CaMn03. vi vi

Manganese trioxide, Mn03 ; acidic, forming manganates, K2Mn04.

Manganese heptoxide, Mn207 ; acidic, forming permanganates,

VII

KMn04.

Manganous compounds. — The manganous salts, MnX2, are
derived from bivalent manganese, and in solution yield the pale pink
cation, Mn" . In the solid state they are pink when water of crystal-
lisation is present.

Manganous chloride, MnCl2. — This salt may be obtained from the
residues after the preparation of chlorine from pyrolusite and hydro-
chloric acid (p. 223): Mn02 + 4HC1 = MnCl2 + C12 + 2H20.
Since pyrolusite always contains ferric oxide, the solution is yellow,
and contains ferric chloride, FeCl3 ; this prevents the crystallisa-
tion of the manganous chloride on evaporation. In order to separate
the iron, one -tenth of the filtered solution which has been evaporated
to drive off excess of acid is precipitated with sodium carbonate.
Ferric hydroxide and manganous carbonate, MnC03, are thrown down.
The precipitate is washed, and added to the remainder of the solu-
tion. On boiling, the whole of the iron is precipitated as ferric
hydroxide, manganese going into solution, and the filtered solution
on evaporation deposits pink monoclinic crystals of MnCl2,4H20 :
2FeCl3 + 3MnC03 + 3H2O = 2Fe(OH)3 + 3MnCl2 + 3C02.

A hydrate MnCl2,6H2O, is formed at — 2° ; at 60°, the ordinary
form of MnCl2,4H2O passes into a second monoclinic form. At 57-85°,
MnCl2,2H2O is obtained, which at 198° gives rose-red, anhydrous
MnCl2. The latter fuses at 650° and volatilises at a higher tempera-
ture ; the vapour density is normal.

Manganous carbonate, MnC03. — By adding sodium carbonate to a
solution of a manganous salt, a pale buff-coloured precipitate of

XLVII MANGANESE 963

manganous carbonate, MnCO3, is formed, which is soluble in water
containing carbon dioxide to form a bicarbonate, and readily
oxidises in air when moist to brown manganic hydroxide,
Mn(OH)3 (cf. FeC03). It occurs in the bright red mineral manganese
spar, isomorphous with calcite ; the mineral manganocalcite,
(Mn,Ca,Mg)C03 is isomorphous with aragonite.

Manganous oxide, MnO. — By heating the carbonate (or any higher
oxide of manganese) in hydrogen, manganous oxide, MnO, is ob-
tained as a greyish -green powder. If the hydrogen contains a trace
of HC1, emerald-green crystals of the oxide MnO are formed.
Manganous oxide is also formed on heating the oxalate :
MnC204 = MnO -f CO -f C02. If a caustic alkali is added to a
solution of a manganese salt, a white precipitate of manganous
hydroxide, Mn(OH)2, is thrown down, which in presence of air or
oxygen rapidly oxidises to brown manganic hydroxide, Mn(OH)3.
This reaction is utilised in estimating the oxygen dissolved in water ;
the precipitate is dissolved in hydrochloric acid, potassium iodide
added, and the iodine titrated. One c.c. of -ZVyiOI2=0-0008gm. of O2.
Ammonia only slowly precipitates Mn(OH)2 from a solution con-
taining ammonium chloride. Probably complex ions are formed, but
the solution rapidly deposits Mn(OH)3 on exposure to air. The usual
method of precipitating the metals Fe, Al, Cr by NH4C1 + NH4OH,
and then precipitating Mn in the filtrate with (NH4)HS, is not applicable
if the latter metal is present in large amounts.

Manganous sulphide, MnS. — This occurs as the mineral alabandite.
It is formed as a grey mass by heating the carbonate with sulphur,
or as a light flesh-coloured, amorphous powder by precipitating a
manganous salt with ammonia and ammonium sulphide. In contact
with excess of ammonium sulphide, it passes into a green, crystalline
form.

The flesh-coloured form is said to be a mixture of a grey and a red
form ; if precipitated with sodium sulphide, the grey form is absent,
and the precipitate does not become green in contact with excess of
reagent.

Manganous sulphide dissolves readily in dilute acids, even acetic ;
in this way manganese may be separated from zinc, the sulphide of
which is insoluble in acetic acid.

Manganous sulphate, MnS04. — This salt is obtained from pyrolusite
by heating with concentrated sulphuric acid : 2Mn02 -f- 2H2S04 =
2MnS04 -f- 2H20 -f- O2. The residue is heated to redness to decom-
pose ferric sulphate : Fe2(S04)3 = Fe203 -f 3S03, dissolved in
water, and the filtered solution evaporated. The last traces of
iron may be removed by boiling with a little manganous carbonate.
The salt forms a number of hydrates : below 8°, MnS04,7H2O,
isomorphous with FeS04,7H2O ; at 8-27 °, MnS04,5H20, isomorphous

3 Q 2

964 INORGANIC CHEMISTRY CHAP.

with CuS04,5H2O ; above 27°, MnS04,H2O. A labile hydrate,
MnS04,4H2O, separates out at 30°. Manganous sulphate forms
well-crystallised double salts, e.g., K2S04,MnS04,6H2O isomorphous
with ferrous ammonium sulphate, (NH4)2S04,FeSO4,6H2O.

Manganous ammonium phosphate, MnNH4P04,H20. — This is
formed as a reddish- white, glittering, crystalline precipitate by the
addition of ammonium chloride, ammonia, and sodium phosphate to
a manganous salt. On ignition, it forms the pyrophosphate, Mn2P;i07.
This is used in the estimation of manganese.

Manganese carbide, Mn3C, is formed from the dioxide and excess of
carbon in the electric furnace. With water, it yields hydrogen and
methane : Mn3C + 6H2O = CH4 + Ha + 3Mn(OH)2.

Manganous oxalate, MnC2O4,2H2O, is obtained as a white, crystalline
precipitate. It loses water at 25°.

Manganese borate, MnH4(BO3)2,H2O, is formed as an almost white
powder by precipitating manganous sulphate with borax and drying
at 100°. At a red heat it forms the metaborate, Mn(BO2)2. The pre-
cipitate is used as a drier, for promoting the oxidation of linseed oil in
paints and varnishes : it acts catalytically, probably by the inter-
mediate formation of a higher oxide.

Manganic salts, MnX3. — Manganic oxide, Mn2O3, occurs in the
mineral braunite ; the hydroxide, Mn(OH)3, occurs in the partly
dehydrated form as manganite, MnO(OH) ; it is formed as a brown
precipitate by passing chlorine through water containing manganous
carbonate in suspension. With hot nitric acid it forms man-
ganous nitrate and manganese dioxide : 2MnO(OH) -J- 2HN03 =
Mn(NO3)2 + Mn02 + 2H20. Manganic sulphate, Mn2(S04)3, is
formed as a dark green powder by heating the precipitated dioxide
with concentrated sulphuric acid at 138°, draining on a porous tile,
washing with concentrated nitric acid, and heating at 150°. It
dissolves in water to a violet liquid, which deposits brown hydrated
oxide on dilution. It forms alums, e.g., K2S04,Mn2(S04)3, 24H2O.

Manganic phosphate, MnP04,2H20, is formed as a greenish-grey
precipitate when a solution of manganous sulphate containing acetic
and phosphoric acids is oxidised by potassium permanganate at
100°. It is insoluble in water, but dissolves in concentrated sul-
phuric or phosphoric acid to form violet solutions. A violet solution
is also obtained by heating a manganous salt with phosphoric and
nitric acids at 150° ; a lilac precipitate of the acid pyrophosphate,
MnHP207, is also formed. Manganese salts give a violet microcosmic
salt bead.

Manganic chloride, or manganese trichloride, MnCl3, is probably
contained in the dark brown solution formed when manganese
dioxide is dissolved in cold concentrated hydrochloric acid :
2Mn02 + 8HC1 = 2MnCl3 -f 4H20 + C12. On warming, chlorine is

XLVII MANGANESE 965

evolved : 2MnCl3 = 2MnCl2 + C12. The dark brown solution
probably also contains the tetrachloride, MnCl4 : MnO2 + 4HC1 =
MnCl4 + 2H20. Crystalline double salts of these two higher
chlorides are known, e.g., MnCl3,2KCl and MnCl4,2KCl. If man-
ganese dioxide is suspended in carbon tetrachloride and dry hydrogen
fchloride passed through, a solid containing MnCl3 and MnCl4 is
formed. If the solid is washed with dry ether, a violet solution of
manganese trichloride, MnCl3, is obtained, and a reddish-brown
powder of manganese tetrachloride, MnCl4, remains. The tetra-
chloride forms a red solution in absolute alcohol. Both higher
chlorides of manganese are decomposed by water, and the dark brown
solution of manganese dioxide in hydrochloric acid also deposits
a brown precipitate when poured into water : MnCl3 -f 3H«O ^
Mn(OH)3 + 3HC1.

Manganese tetrafluoride, MnF4, on the other hand, is not decom-
posed by water. It is obtained by dissolving the dioxide in hydro-
fluoric acid, and forms double salts, e.g., K2MnF6.

Mangano-manganie oxide, Mn304. — This oxide, known as red
oxide of manganese, occurs in the mineral Tiausmannite. It is formed
when any other oxide of manganese is heated strongly in air :
3MnO + 0 = Mn304 ; 3Mn02 = Mn?O4 + 02. It dissolves in
cold concentrated sulphuric acid, forming a red solution containing
manganous and manganic sulphates : Mn304 + 4H2S04 = MnS04 +
Mn2(SO4)3 + 4H20. Acetic acid gives a solution of manganous
acetate and a residue of manganese sesquioxide, Mn2O3, hence the
red oxide may be regarded as a mixed oxide, MnO,Mn2Os, analogous
to red lead, or as manganous manganite, Mn(Mn02)2.

Manganese dioxide, Mn02. — This oxide occurs native as pyrolusite.
It is prepared in the pure state by heating manganous nitrate until
red fumes appear, decanting the clear liquid from the residue of
lower oxides, and heating it for forty to sixty hours at 150-160°.
If solutions of manganous salts are treated with oxidising agents
such as potassium permanganate, sodium hypochlorite, ammonia
and bromine, or ozone, brown precipitates are obtained, which on
washing form brown colloidal solutions. These precipitates,
however, always contain less oxygen than corresponds with the
formula Mn02. Manganese dioxide is a feebly acidic oxide, and
with strong bases forms salts called manganites, e.g., CaO,MnO2
and CaO,2Mn02 (see Mn3O4).

The commercial dioxide is used as an oxidising agent, and may be
analysed as follows : ( 1 ) The solid is boiled with a standard solution
of oxalic acid, containing sulphuric acid. A portion of the oxalic
acid is oxidised: C2H2O4 + MnO2 -f H2SO4 = 2CO2 + MnSO4 + 2H2O.
The excess of oxalic acid is then titrated with potassium permanganate
solution. (2) The oxide is heated with concentrated hydrochloric acid

966 INORGANIC CHEMISTRY CHAP.

in a small flask, and the chlorine evolved passed into a solution of
potassium iodide contained in a U-tube cooled by water. Iodine is
liberated, which is titrated with standard sodium thiosulphate solution.
(3) The dioxide is boiled with an acidified solution of standard ferrous
sulphate in a flask fitted with a tube dipping under water to exclude air.
The excess of ferrous sulphate is titrated with standard permanganate
solution : MnO2 + 2FeSO4 + 2H2SO4 = MnSO4 + Fe2(SO4)3 + 2H2O
The first method usually gives the most accurate results.

Manganic disulphate, Mn(SO4)2, corresponding with manganese
dioxide, is obtained in black crystals or a deep brown solution by the
electrolysis of a solution of manganous sulphate in fairly concentrated
sulphuric acid with a platinum anode. It is used as an oxidising agent.

Besides its use in decolorising glass, manganese dioxide is applied,
mixed with ferric oxide, as a dark brown glaze to pottery. It is
also used as a depolariser in the Leclanche cell.

In the simplest form this consists of a rod of amalgamated zinc
immersed in a concentrated solution of ammonium chloride, in which is
also placed a porous pot containing a rod of gas carbon surrounded
by a granular mixture of crushed pyrolusite and gas carbon. In a
second form, the pyrolusite is formed into blocks, one of which is placed
on each side of a gas-carbon plate, being held in position by rubber
bands. In the dry cell, used in enormous numbers for portable lamps and
other purposes, the ammonium chloride solution is gelatinised by adding
glue, the gas-carbon rod is surrounded by a gelatinised paste of man-
ganese dioxide and ammonium chloride, and the zinc pole consists
of a zinc cylinder containing the materials of the cell.

The reaction in the cell is the solution of zinc to form a double
chloride : Zn + 5NH4C1 = ZnCl2,3NH4Cl -f 2NH3 -f H2. The hydrogen
is deposited on the pyrolusite, and is oxidised by the trace of man-
ganic ions formed by the minute amount of the manganese dioxide
in solution :

MnO2 (solid) -f Aq. ±^ Mn(OH)4 (dissd.) ^± Mn" + 4OH'.

Mn" + H + OH' n Mn" -f H2O.

Mn'" + 3OH' ^± Mn(OH)3 (dissd.) ^ Mn(OH)3 (ppd.)

The compound ZnCl2,3NH4Cl slowly separates in crystals on the
zinc rod.

The cell rapidly polarises, since the concentration of depolarising
Mn:: ions is small, but recovers fairly quickly on standing, when more
MnO2 goes into solution. It is useful when intermittent currents of
short duration are required, as in operating bells or flash-lamps.

Manganates and permanganates. — If manganese dioxide is fused
with caustic soda or potash with free access to air, a green mass
is formed which contains a manganate, e.g., K2Mn04. The reaction
is more complete with caustic potash (2 4 mols. to 1 mol. of Mn02) and

XL VII

MANGANESE 967

more rapid if potassium or sodium nitrate or chlorate is added to
the alkali : 4KOH + 2Mn02 + O2 == 2K2Mn04 + 2H2O. The
dark green mass may be dissolved in a small quantity of cold water,
forming a dark green solution, from which on evaporation in a vacuum
dark green crystals of the manganates, K2MnO4, or Na2MnO4,10H2O,
are deposited. These are isomorphous with the corresponding
sulphates, K2S04 and NaS04,10H20. Sodium manganate is used
as a disinfectant, since it is a powerful oxidising agent.

If the dark green solution of the manganate in a little water is
poured into a large volume of water, a purple solution of perman-
ganate and a brown precipitate of hydrated manganese dioxide are
formed : SKaMnC^ + 2H2O = 2KMn04 + 4KOH + Mn02. In
presence of a large excess of alkali, the reaction does not take place,
and the manganate is stable. The reaction occurs completely if
the alkali produced is removed by adding an acid ; even carbonic
acid is effective. If a pure alkali is added to the purple solution of
permanganate, no reaction occurs, but as commercial alkali always
contains nitrites, which are readily oxidised, this causes the colour
to change again to green : 2KMn04 + 2KOH = 2K2MnO4 + H2O
+ 0. With very concentrated solutions of permanganate and pure
alkali, however, this reaction occurs spontaneously, and oxygen gas is
evolved.

The formation of manganates and permanganates by the above
reactions was discovered by Glauber in 1656 ; on account of the
colour changes which it undergoes the manganate was called mineral
chamelion by Scheele. The salts were investigated by Forchhammer
in 1820 and by Mitscherlich in 1832. The latter showed that the
green and purple salts were derived from two distinct acids, man-
ganic acid, HjjMnC^, and permanganic acid, HMn04, and that the
salts are isomorphous with sulphates and per chlorates, respectively.
Potassium permanganate may be obtained from the manganate by
passing chlorine through the solution : 2K2Mn04 -f- C12 =
2KMn04 + 2KC1.

Permanganic acid, HMn04. — Manganic acid is not known in the
free state, since manganates, when treated with other acids, do
not give manganic acid but permanganates. Permanganic acid,
HMn04, is obtained in solution by boiling a solution of manganous
sulphate with lead dioxide and nitric acid. If a solution of silver
nitrate and potassium permanganate is crystallised, silver permanganate,
AgMn04, is obtained. If this is decomposed with barium chloride,
barium permanganate, Ba(Mn04)2, is obtained, which, when treated with
dilute sulphuric acid, gives a purple solution of permanganic acid,
violet crystals of which are formed by evaporation in a vacuum. The
acid is unstable ; the solution rapiflly decomposes with evolution
of oxygen and deposition of manganese dioxide : 4HMn04 =
4Mn02 + 2H20 + 302. It is a powerful oxidising agent.

968 INORGANIC CHEMISTRY CHAP.

Manganese heptoxide, or permanganic anhydride, Mn207. — When
powdered potassium permanganate is added in small quantities
at a time to cooled concentrated sulphuric acid, a dark green solu-
tion is formed, which appears to contain the sulphate of manganese
trioxide, (Mn03)2S04, or Mn207,SO3. This green liquid is liable
to explode violently in contact with traces of organic matter, or
even spontaneously. When treated with ice-cold water, dark
oily drops of manganese heptoxide, Mn207, the anhydride of per-
manganic acid, separate :

2KMnO4 + 2H2S04 = (Mn03)2S04 + KoS04 + 2H2O.
(Mn03)2SO4 + H20 = Mn207 + H2S04.

Manganese heptoxide is an opaque, oily liquid, sp. gr. 24, which
forms a violet vapour at 40-50°, but explodes violently on warming
or in presence of organic matter. It dissolves unchanged in glacial
acetic acid.

If fused sodium chloride is added to the green solution of potass-
ium permanganate in concentrated sulphuric acid, a yellow gas
is evolved which condenses in a freezing mixture to a greenish-
brown liquid, permanganyl chloride, Mn03Cl, the acid chloride of
permanganic acid. It explodes on heating, and in moist air emits
purple fumes, owing to hydrolysis into hydrochloric and perman-
ganic acids. The same reaction occurs in presence of water, but
the two acids mutually decompose each other, with formation of
hydrochloric acid and manganese dioxide. The corresponding
fluoride, Mn03F, has been prepared.

The oxide MnO3 has been described, but its existence is doubtful.
It is said to be formed by dropping the green solution of KMnO4 in
H2SO4 on dry sodium carbonate, but the purple fumes evolved are
more probably permanganic acid droplets, formed from the water
produced by the interaction of H2SO4 and Ma2CO3.

Potassium permanganate, KMn04. — This important salt is made
by fusing manganese dioxide with caustic potash and potassium
nitrate or chlorate, boiling the green mass of manganate with water,
filtering the solution through asbestos, and passing a current of
carbon dioxide through the solution. A purple solution of potassium
permanganate is formed : 3K2Mn04 +2H20 + 4C02 = 2KMn04 -f
MnO<> -j- 4KHC03. The solution is again filtered through asbestos
or glass-wool, and evaporated. Deep purple-red, brilliant rhombic
prisms of the permanganate separate. These have a green irides-
cence and dissolve in water (44 in 100 at 10° ; 5-31 at 15° ; 324
at 75°) to a deep purple solution, which is opaque unless dilute.
The crystals evolve oxygen on heating and fall to a black powder :
2KMnO4 = K2Mn04 -f Mn02 -f O2. At a red heat the manganate
is also decomposed, with evolution of oxygen.

XLVII MANGANESE 969

Potassium permanganate is also manufactured from the man-
ganate by the electrolytic oxidation of the solution between iron
or nickel electrodes. If an electrode of manganese or ferroman-
ganese is made the anode in a solution of caustic potash, and a
nickel cathode used, a solution of the permanganate may be obtained
directly, but the yield is small.

Calcium permanganate, Ca(Mn04)2, is manufactured by the addi-
tion of chalk to a solution of permanganic acid obtained by elec-
trolytic oxidation of a manganous salt. It is a deep violet, hygro-
scopic powder, readily soluble in water, and is used in sterilising
water. It loses oxygen more readily than the potassium salt.

Potassium permanaganate is a powerful oxidising agent. It burns
violently when mixed with sulphur or charcoal and ignited. The
oxidising action is different according as the reaction is carried out
in alkaline or in acid solution.

(1) In alkaline solution, in the presence of reducing agents, the
permanganate is first reduced to green manganate. The solution
then deposits brown manganese dioxide and becomes colourless.
The reactions are :

2KMn04 + 2KOH = 2K2Mn04 + H2O + O.
2K2MnO4 + 2H2O = 2MnO2 -f 4KOH + 20.

Hence, two molecules of potassium permanganate in alkaline
solution give three atoms of available oxygen when reduced to man-
ganese dioxide. The reaction may also be represented in the
dualistic notation (p. 275) as follows :

2KMnO4 K2O,Mn207 = K2O + 2Mn02 + 30.

EXPT. 334. — Add a little sugar solution to an alkaline solution of
potassium permanganate and warm. Observe the change of colour to
green, followed by the discharge of colour and the formation of a brown
precipitate.

Alkaline permanganate oxidises an iodide to an iodate :
2KMn04 + H20 + KI 2MnO2 + 2KOH + KI03.

A manganous salt is oxidised in neutral solution to manganous
manganite :

4KMn04 -f HMnS04 + 14H20 =

4KHS04 + 7H2S04 -f 5Mn(HMn03)2.

(2) In acid solutions, the permanganate is reduced to a man-
ganous salt and~jfa>e atoms of oxygen become available from two
molecules of permanganate :

2KMn04 + 3H2S04 = K2SO4 + 2HMnO4 + HoO + 2H2S04.

= K2S04 -f 2MnS04 + 3H2O + 5O ;
or, in dualistic notation :

2KMn04 = K20,Mn207 = K2O + 2MnO -f 5O.

970 INORGANIC CHEMISTRY CHAP.

In acid solutions, iodine is liberated from potassium iodide :

2KMnO4 + 10KI + 8H2SO4 = 6K2SO4 + 2MnSO4 + 5I2 + 8H2O.

Ferrous salts are oxidised to ferric salts :

2KMnO4 + 10FeSO4 + 8H2SO4 =

K2SO4 + 2MnSO4 + 5Fe2(SO4)3 + 8H2O.

Oxalic acid is oxidised to carbon dioxide :
2KMnO4 + 5C2H2O4 + 3H2SO4 = K2SO4 + 2MnSO4 + 10CO2 + 8H2O.

Nitrites are oxidised to nitrates :
2KMnO4 + 5HNO2 + 3H2SO4 = K2SO4 + 2MnSO4 + 3H2O + 5HNO3.

Sulphur dioxide is oxidised to sulphuric acid :

2KMnO4 + 5SO2 + 2H2O = K2SO4 + 2MnSO4 + 3H2SO4.

The reaction with hydrogen peroxide has been described (p. 162).
The reactions are accelerated by the presence of manganous salts, which
act catalytically.

EXPT. 335. — To a solution of oxalic acid acidified with sulphuric
acid and warmed at 60°, add potassium permanganate solution from a
burette. With the first few c.c. the colour is discharged only slowly,
but as manganous sulphate accumulates, the colour is quickly
discharged.

In volumetric analysis, solutions of potassium permanganate are
made up according to the content of available oxygen. A normal
solution is one containing one gram equivalent of active substance per
litre. In the case of permanganate this will be 7-94 gm. of available
oxygen. The solubility of the salt is not sufficient to give a normal
solution, so that semi-normal (N/2) and decinormal (N/10) solutions
are used. 2KMn04 give 50 .*. a normal solution will contain
2KMn04 2KMn04

—= ^gm., and a decinormal solution -^ y o x 10 = 3-137 gm. per

litre. The solution may be standardised by oxalic acid. In the
oxidation of ferrous salts, 2FeO require 30 to form Fe203, or 554 gm.
of iron require 7-94 gm. of oxvgen : hence 1 c.c. of Ar/10KMn04 =
0-00554 gm. of Fe.

Cyanogen compounds. — Potassium cyanide gives with solutions of
manganous salts a yellowish-grey precipitate of manganous cyanide,
Mn(CN)2. This is soluble in excess of the reagent, giving a yellow
solution of potassium manganocyanide, K4Mn(CN)6, analogous to
the ferrocyanide, which crystallises as a deep blue solid,
K4Mn(CN)6,3H20. By evaporating this solution in air, a portion
of the manganese is oxidised and precipitated, and the solution
contains potassium manganicyanide, K3Mn(CN)6. This forms large
red prisms. The resemblance between manganese and iron is
apparent : the corresponding salts are isomorphous.

XLVII MANGANESE 971

EXERCISES ON CHAPTER XLVII

1. Discuss the position of manganese in the Periodic System. What
analogies does it show to iron and chromium ?

2. How are the oxides of manganese prepared, and what are their
properties ? How is the available oxygen in manganese dioxide
estimated ?

3. Describe the preparation of a pure manganous salt from pyro-
lusite.

4. How is manganous sulphide prepared, and what are its pro-
perties ?

5. Describe the preparation of potassium permanganate. Give
examples of its action as an oxidising agent. How would you determine
the strength of a solution of hydrogen peroxide by means of potassium
permanganate ?

6. How are the following obtained : (a) manganese trichloride,
(b) potassium manganocyanide, (c) manganese pyrophosphate, (d)
manganese heptoxide ?

CHAPTER XLVIII
EBON

The transitional elements. — The eighth group of the Periodic
System comprises three sub-groups, with three elements in each,
forming the termination of the even series 4, 6, and 8, and con-
necting the elements of these series with those of the odd series
following. For this reason they were called by Mendeleeff the
transitional elements :

Group A: Iron, 55-40; Cobalt, 58-50; Nickel, 58-21.

Group B: Ruthenium, 100-9; Rhodium, 102-1; Palladium, 105-9.

Group C: Osmium, 189-4; Iridium, 191-6; Platinum, 193-6.

The propriety of separating these elements from the rest and
placing them in a special group is justified by their peculiar pro-
perties. The atomic weights of the members of each of the three
groups differ so little from one another that the three elements may
be regarded as forming a single cluster which takes the place of a
single element in the other groups of the Periodic System. A similar
behaviour is shown by the rare-earth elements (p. 461).

The physical and chemical properties of the elements are also
closely related ; e.g., the platinum metals of groups B and C are
very similar ; they are difficult to separate, as are also cobalt and
nickel. The elements in the vertical columns show close resem-
blances ; ruthenium and osmium form higher oxides, R04 ; rho-
dium and iridium, palladium and platinum also exhibit analogies.
The resemblance between the metate of the iron group and the
platinum metals is, however, somewhat remote, and is chiefly
confined to the facility with which all the metals of Group VIII
form complex compounds :

potassium ferrocyanide, potassium cobaltinitrite,

K4Fe(CN)6; KgCotNO^;

potassium chloroplatinate, sodium osmichloride,

K2PtCl6; Na2OsCl6,2H20

Nickel shows this property to a much smaller degree. The
platinum metals, both in their physical properties, e.g., their
" noble " character, and their tendency to complex-formation,

972

CH. XLVIII IRON 973

closely resemble gold, which follows them in the periodic table.
All the elements of Group VIII, unlike the other members of even
series, form organo -metallic compounds.

The typical oxide of these elements should be, according to
their position in the Periodic System, RO4, but this is confined
to ruthenium and osmium. All the elements form lower, basic
oxides.

The elements of the iron group : iron, cobalt, and nickel, are mag-
netic metals with high melting points, which oxidise in the air at
a red heat, and decompose steam at high temperatures. The oxides

ii in

RO are all known, and strong bases. The sesquioxides, R203, are
also basic, but their salts are stable only in the case of iron. Oxides
of the type R3O4 are also known ; their salts, if they exist, are very

unstable, so that these oxides probably have the formula RO, R203,
or R(R02)2, in which R203 functions as a feebly acidic oxide, and
RO as a base. Compounds such as CaO,Fe203, or Ca(Fe02)2
(ferrites), are known. Iron forms compounds of an unknown acidic
trioxide, Fe03, e.g., potassium ferrate K2Fe04, in which the element
shows a resemblance to manganese and chromium, which form
K2Mn04 and K2CrO4. The metals iron, manganese, and chromium
are also similar in their physical properties.

The elements of Group VIII all form compounds, called carbonyls
with carbon monoxide : e.g.. Ni(C04). Molybdenum forms a
carbonyl. In these compounds the metal appears to function with
its maximum valency :

CO^ /.CO

IRON. Fe = 5540.

Iron. — The element iron, by reason both of its abundant occur-
rence and of its manifold uses, is undoubtedly the most important
of the metals. On account of its high melting point and the
comparative difficulty with which iron is reduced from its ores,
the metal was probably not known until a later period (c. B.C. 1500)
than bronze (c. B.C. 3000), although the more permanent character
of the latter metal may be the reason why the bronze implements
of prehistoric man have persisted, whilst iron, if it existed at all,
has rusted away and disappeared. It was certainly regarded by the
ancients as a rarity, since Homer refers to the prize of a ball of iron
awarded to Achilles for his athletic skill. From its obvious associa-
tion with military exploits, the alchemists named the metal after

974 INORGANIC CHEMISTRY CHAP.

the planet Mars and denoted it by the symbol of the spear and
shield : £ .

Iron does not occur to any great extent in the free state on the
earth, although meteorites, which consist of metallic iron with from
3 to 30 per cent, of nickel, indicate that it must be present in the
solar system. Meteorites may consist partly of silicates (e.g.,
olivine), and of glassy minerals (moldavite). On account of the
presence of nickel, meteoric iron does not easily rust in moist air.
Meteoric dust, consisting chiefly of iron,, is constantly foiling on the
earth from space, although its presence is only noticed on the
surface of the otherwise unsullied snows of the polar regions. Large
masses of native iron, which may be of meteoric origin, or have been
derived from the reduction of ores in burning coal-mines, occur in
many localities, particularly at Disko Island. West Greenland.
Metallic iron also occurs in grains in basalt rocks, found at Giant's
Causeway, and elsewhere. Iron is contained in the chlorophyll of
green plants, and in haemoglobin (0*336 per cent. Fe), the red
colouring matter of blood.

Iron ores. — The ores of iron are plentiful but relatively few in
number, although the element occurs in nearly every mineral, in
rocks, and in soils. The most important ores are the oxides. The
black oxide, Fe304, or ferroso-ferric oxide, occurs as the important
ore magnetite, so-called because certain varieties (lodestone) are
permanently magnetic. This ore is not found to any extent in the
British Isles, but occurs in Lapland, Sweden, Siberia, Germany,
and North America. It contains 724 per cent, of iron, and is
the richest ore of the metal. The sesquioxide, Fe203, occurs as
hcematite, which is crystalline and has a red colour, or if black,
as is sometimes the case, gives a red streak when drawn
over an unglazed porcelain plate. It is found in Belgium,
Sweden, the Island of Elba, south of Lake Superior, and in
England in the Furness district in Lancashire. The hydrated
sesquioxide, 2Fe203,3H2O, limonite, occurs in kidney-shaped
amorphous masses in South Wales, the Forest of Dean, and at Bilbao
in Spain. The so-called bog iron ores are hydrated oxides, and occur
in large quantities in Ireland, Sweden, and North Germany. The
only remaining important ore is ferrous carbonate, FeC03, occurring
either alone, as siderite, chalybite, or spathic iron ore, or mixed with
clay as clay -ironstone, or with clay and coal as blackband-ironstone.
The hydrated oxide and the impure forms of the carbonate are the
most important British ores.

Pyrites cinders, chiefly ferric oxide, from the manufacture of
sulphuric acid (p. 503) are desulphurised by roasting and smelted
for iron.

The value of an ore of iron depends on its freedom from impurities
(S, P, As, etc.), which are detrimental to the resulting metal. Three

IRON

975

XLVIII

varieties of commercial iron are made : (1) cast-iron, or pig-iron ;
(2) malleable iron, or wrought-iron ; (3) steel.

The metallurgy of iron. — The order in which the varieties of iron
are prepared from the ore is roughly as follows :

p Wrought-iron -> Crucible Steel.
Ore ~> Cast, or Pig-iron

^Bessemer, or Open-hearth Steel.

The extraction of iron
from the ore involves a
number of processes.

(1) preliminary roasting,
or calcination, to drive
off carbon dioxide and
moisture and leave ferric
oxide, Fe203. This opera-
tion is carried out by
stacking the ore with a
little coal in heaps or in
shallow kilns, and regu-
lating the temperature
and supply of air so that
most of the moisture,
carbon dioxide, sulphur,
and arsenic are expelled ;
ferrous oxide (FeO) is also
converted into ferric
oxide (Fe203), to avoid
the production of ferrous
silicate in the slag during
smelting. The ore is also
rendered more porous.

(2) Smelting, or reducing
the ore with carbon in
the blast-furnace.

The blast - furnace. —
The blast-furnace, intro-
duced about 1500 (Fig.
420) consists of an outer
shell of steel plates, lined
with refractory bricks.
It is 50-100 ft. high, the
greatest width being
about 24 ft. (at the

boshes"). The mouth

4P Fee*

FIG. 420. — Blast Furnace.

uu»ue» ). JLXIC uiuuvii is closed with a cup-and-cone, B,
through which a mixture of ore and fuel is fed intermittently

976 INORGANIC CHEMISTRY CHAP.

into the furnace, whilst the gases (carbon monoxide and nitrogen)
pass away through the pipe, F, to a dust-catcher, G, and are utilised
in heating the blast. The furnace below the boshes narrows gradually
to a hearth, C, at the base, about 10 ft. in diameter, and the same
height. This is pierced with holes for the water- jacketed iron
blowing-pipes, or tuyeres, through which air is forced from the
annular pipe, H, by means of powerful blowing engines. The
hearth is also pierced with a hole, E, from which the molten iron is
periodically tapped into sand moulds on the ground, and a slag-notch
(not shown in the figure) at a higher level, through which the
molten slag runs continuously from above the fused metal. About
3-4 tons of air are passed through the furnace per ton of iron made,

the power for work-
ing the blowing-
engines being sup-
plied by coke-oven
gas obtained in pro-
ducing the coke for
the blast furnace.

The charge for the
blast-furnace con-
sists of 1 ton of
hard oven- coke and
&-12 cwt. of lime-
stone (to form the
slag, consisting of
calcium and alum-
inium silicates) to
so much ore (say
TO chimney 2J tons) as produces
1 ton of iron. The
process is conthi-
uous and goes on
for years without interruption. The furnace should not be allowed
to cool, when a hard mass of slag and metal would be produced,
which has to be blasted out and the furnace re-started, an operation
lasting some months. Each furnace may produce 1000-1300 tons
of cast-iron per week.

The air blast has since 1828 been pre-heated to 700-800°. This
hot blast is now produced by passing the air through Cowper stoves
(Fig. 421), consisting of tall iron cylinders lined with firebricks,
packed on one side with chequer brickwork. Part of the hot gas
from the blast furnace, together with sufficient air to burn it, passes
through these until the bricks are heated to redness. The gas is
then turned through a second stove and the air blast to the tuyeres
sent through the first one until the brickwork has cooled. The two

Cold blast

FIG. 421.— cowperstove.

XLVIII IRON 977

stoves are thus alternately used as absorbers and emitters of heat, or
as heat-regenerators. In this way an economy of fuel is effected,
and the furnace works at a higher temperature. The normal com-
position of blast-furnace gas, by volume, is : N2, 60 ; CO, 24 ;
CO, 12 ; H2 and CH4, 4.

In some cases a dry blast is used, the air passed through the Cowper
stoves being first cooled by refrigeration to remove moisture. In this
way loss of heat by the reaction : C + H2O ^ CO -f H2, in the blast
furnace, is said to be prevented. The furnace gases, after cooling by
passing through long iron pipes sprayed with water, are filtered through
cloth bags, or treated by electrostatic precipitation, to remove dust,
which may be rich in potassium salts. If coal is used in the
furnaces the cooled gases are scrubbed with water to recover the
ammonia.

Chemical reactions in the blast furnace. — The oxygen of the blast
unites with carbon in the hearth to produce carbon monoxide :
(1)2C -f- O2 ^ 2CO. The temperature of the charge passing down
the furnace increases continually from the mouth to the hearth.
The reactions in different parts of the furnace, starting at the mouth,
will now be considered.

Above the boshes, at a dull red-heat (500-900°), the ferric
oxide is reduced by the carbon monoxide to spongy iron :

Fe203 + SCO =z= 2Fe + 3C02.

The reaction is reversible, and the escaping gases contain CO and
CO2 in the ratio 1 to 0-5. Two other reactions also occur, which
limit the completeness of the reduction : 2Fe -f CO2 ^ Fe20o +
CO, and 2Fe + SCO ^± Fe2O3 + 30.

In this upper zone the limestone is decomposed : CaC03 ^±
CaO -f- C02, and some carbon dioxide is reduced to monoxide :
CO -(- C ^± 2CO. The spongy iron absorbs sulphur from the fuel.

Near the centre of the furnace, at a bright red heat, finely-divided
carbon is deposited by the reaction : 2CO ^± CO2 + 0 ; phosphorus
is produced by reduction of phosphates in the ore : P2O5 + 5Fe -f-
5Si02 = SFeSiOg + 2P, and aUoys with the iron. At a higher tem-
perature silicon is formed by reduction of silicates with iron and
carbon, and alloys with the iron, whilst a portion of the silica unites
with bases (CaO,Al203) to form a fusible slag.

At a white heat in the lowest part of the furnace the spongy iron,
containing carbon, sulphur, phosphorus, and silicon, fuses to molten
cast-iron, which is tapped off from time to time into sand moulds
to form pig-iron, or is sent in the fused state to the steel furnaces.
The heat evolved hi the main reaction is : Fe003 + SCO = 2Fe -f-
3C02 + 8-65kg. cal.

Cast-iron. — Pig-iron contains from 3 to 4 per cent, of carbon,
together with silicon, sulphur, phosphorus, and manganese. When

3 B

978 INORGANIC CHEMISTRY CHAP.

the cooling is rapid, the silicon content small, and the manganese
high, white pig- iron is formed, in which all the carbon is in the form
of iron carbide, Fe3C (cementite). It is brittle, coarsely crystalline,
and dissolves nearly completely in dilute hydrochloric acid, evolving
a mixture of hydrogen and hydrocarbons. If, however, the molten
iron containing at least 2-5 per cent, of silicon is slowly cooled,
most of the carbon separates in the form of fine laminae of graphite,
the metal at the same time becoming softer and of a finer texture ;
on solution in hydrochloric acid it evolves chiefly hydrogen and leaves
a black residue of graphite. This variety of cast-iron is known as
grey pig-iron. An intermediate variety is called mottled pig-iron.
The solubility of carbon in pure iron is 4-5 per cent. ; much more is
dissolved if manganese is present.

Malleable, or wrought, iron. — This variety is nearly pure iron,
containing only from 0-12 to 0-25 per cent, of carbon, and melts
at a higher temperature (1400-1500°) than cast-iron. Malleable
iron contains less than 0-5 per cent, of total impurities (carbon,
sulphur, phosphorus, silicon).

Malleable iron is obtained from cast-iron by the puddling process,
invented by Cort in 1784. The cast-iron is fused in a reverberatory
furnace, the hearth of which is lined with haematite, which oxidises
the carbon : 30 + Fe203 = 2Fe -f- SCO, the carbon monoxide
bubbling through the molten iron. Sulphur, phosphorus, and
silicon are oxidised and pass into the slag. When the metal becomes
pasty it is formed into lumps, or " blooms," which are beaten under
steam hammers to squeeze out the slag.

The iron, although not fused, welds together to a coherent mass
at about 600°. Malleable iron is tough and fibrous ; its property
of welding, whereby two pieces when heated to redness unite on
hammering, is exceedingly valuable and is applied in various
ways by the blacksmith. Its softness is not appreciably altered by
heating to redness and quenching in water, whereas steel then
becomes very hard.

If wrought iron contains combined phosphorus, it is brittle at the
ordinary temperature, and is said to be cold-short ; combined sulphur,
probably FeS, renders the metal brittle at a red heat, when it is known
as red-short.

Steel. — Steel is iron which has been fused in the process of manu-
facture and contains 0-15 to 1-5 per cent, of combined carbon
dissolved in the form of cementite, Fe3C. It may also contain
manganese, titanium, chromium, nickel, tungsten, and vanadium.
Steel may be made (1) from cast-iron by removing part of the carbon,
(2) from wrought-iron by adding combined carbon.

Before Cort's discovery, wrought-iron was made from pure oxide
ores by reduction with charcoal and was converted into steel by the

XLVIII IRON 979

cementation process. Bars of wrought-iron are heated with charcoal for
one or two weeks. Absorption of carbon gradually occurs, the carbonisa-
tion spreading slowly through the mass, and converting the iron into
steel. The surface of the bars is covered with blisters, and the " blister
steel " is fused in plumbago crucibles to form cast -steel or crucible
steel. The addition of spiegel, an alloy of iron, carbon, and manganese,
to the molten steel improves its quality. The mechanism of the
absorption of carbon by iron is not very clear. It is stated that pure
carbon, free from gases, does not penetrate iron except under high
pressure, so that carbon monoxide may be the active agent. Unstable
iron carbonyls may be formed as intermediate products (p. 992).

Modern steel is produced by removing part of the carbon of cast-
iron by :

(1) The Bessemer process (Kelly, 1852 ; Henry Bessemer, 1855).

(2) The Open-hearth process (Siemens-Martin process, W. Siemens, 1863,
and E. Martin, 1864).

The Bessemer process. — This process is, after Cort's discovery,
one of the master-processes in the metallurgy of steel. The molten
iron from the blast furnaces is run into a converter (Fig. 422), a
large pear-shaped iron vessel. A, lined with refractory silica bricks,
C. The converter holds 10 tons of metal, and is supported on
trunnions, air being led by a pipe, D, to a hollow perforated bottom,
My from which it is forced through the metal. The charging with
molten cast-iron is carried out through the open mouth with the
converter in a horizontal position, and blowing is then begun. The
converter is next swung into a vertical position, and the blowing
continued. Silicon is first oxidised to silica which passes into the
slag, then a portion of the iron is oxidised. The resulting ferric
oxide removes the carbon, forming carbon monoxide, which is
freely evolved from the molten iron and burns at the mouth of the
converter as an orange-yeUow flame edged with blue, shot through
by showers of sparks. After six to eight minutes the flame sinks,
indicating that the carbon has been removed. The converter is
again tilted, the blast stopped, and the requisite amount of spiegel
added — a method of carburising the metal introduced by Mushet
in 1856. The molten steel is poured, by further tilting the con-
verter, into ladles supported by travelling cranes, from which it is
run into moulds. A little silicon-iron alloy (silicon-spiegel) , or
titanium-iron alloy, may be added to remove blow-holes in the cast-
ings due to bubbles of gas, which combine with the silicon or titanium
(CT2, N2, CO). According to the percentage of carbon added,
various kinds of steel are produced : tool steel (0 -9-1 -5 per cent. C) ;
structural steel (0-2-0-6 per cent. C) ; mild steel (0-2 per cent., or
less, C). Special steels are produced by adding alloys of iron with
tungsten, chromium, molybdenum, and vanadium ; or nickel.

3 R 2

980

INORGANIC CHEMISTRY

CHAP.

The German ores of iron are all phosphatic, and the resulting iron
or steel, if made in the ordinary way, would be cold-short. Such
" phosphatic ores " may be worked by the process of Thomas and

XLVIII

IRON

981

Gilchrist (1879), in which the silica (" acidic ") lining of the converter
is replaced by a " basic " lining of magnesia and lime, prepared by
calcining dolomite. Limestone is first charged into the converter,
along with coke, and the blast turned on. Molten pig-iron is
then run in and the blast continued. Carbon is first burnt out as
usual, but if the blast is prolonged after the flame drops the phos-
phorus is oxidised, unites with the lime and forms a slag containing
calcium phosphate and free lime (basic slag, or Thomas slag), which
is a valuable fertiliser. Spiegel is then added to form the steel.
In this way it is possible to treat pig-iron containing as much as

FIG. 423.— Siemens-Martin Process.

3 per cent, of phosphorus. The steel pigs produced by casting
are annealed in underground furnaces (" soaking-pits ") heated
by blast-furnace gas, and are then passed through the rolling mills
for the production of steel bars.

The open-hearth process. — The open-hearth process is carried out
in a large flat crucible enclosed in a furnace (Fig. 423) heated by
producer gas. The air and gas are supplied through separate
regenerators of chequer brickwork, used in pairs and alternately
traversed by the hot products of combustion and the gases, as in
the case of Cowper stoves (p. 976). Molten cast-iron from the blast-
furnace is run on the hearth, which is lined with ganister in the acid
process or calcined magnesite or dolomite in the basic process. The

982 INORGANIC CHEMISTRY CHAP.

requisite amount of haematite, Fe203, is then added, so that a portion
of the carbon is burnt out of the cast-iron and fluid steel remains.
The subsequent operations are the same as in the Bessemer process.
The furnace may be made to tilt and discharge a portion of its con-
tents into the ladle. The operation lasts 8-10 hours ; it is more
easily controlled than the Bessemer process, and is very largely
used.

Electric furnaces are used in the production of special high quality
steels. They are mostly on the arc principle, and consist of refrac-
tory crucibles containing two (or more) large carbon electrodes
between which an electric arc is struck.

In 1917, the total output of steel was 75,000,000 tons, of which
42,000,000 tons were produced in the United States of America.

The properties of steel. — The properties of steel depend largely
on the content of carbon : low-carbon steels are soft, like wrought-
iron, and are known as mild-steel ; with -further addition of carbon
the ductility falls, whilst the tensile strength increases up to the
limiting percentage of 1-5 C. Cast-iron has a tensile strength of
10 tons per sq. in., wrought-iron of 25 tons, and steel of 30^0 tons.
Wrought-iron and steel are highly malleable and may be welded.
The melting-point of steel is lower than that of wrought-iron. The
properties of steel depend also on the heat-treatment to which the
metal has been subjected. If steel is heated to redness and plunged
into cold water it becomes as hard and brittle as glass. If it is now
heated to various temperatures, the resulting metal possesses pro-
perties depending on the temperature. This operation is known as
tempering, and the temperature is judged by the colour of the thin
film of oxide produced on a bright surface of the metal :

230° : light straw colour : used for razor blades.

255° : brownish-yellow : used for penknives and axes.

277° : purple : used for cutlery.

288° : bright blue : used for watch-springs and swords.

290-316° : dark blue : used for chisels and large saws.

Allotropic forms of iron. — The changes occurring in the tempering
of steels are believed to be the following. There are three allotropic
modifications of iron. (1) a-Ferrite (the chief constituent of wrought-
iron) is stable below 760°, is soft, magnetic, and capable of dissolving
but little iron carbide, Fe3C. (2) /3-Femte is produced at 760°, it
is non-magnetic, and dissolves only a little carbide. (3) y-Ferrite
is produced by heating to 900° ; it is non-magnetic, but differs
from the other two varieties in forming solid solutions with iron
carbide. On cooling, the changes are reversed :

a-ferrite ^± /8-ferrite ^± y-ferrite.

760° 900°

When fluid iron containing dissolved carbon is quickly cooled by

XLVIII IRON 983

quenching, it solidifies to y-ferrite containing dissolved carbide,
Fe3C (cementite) ; the product, which is homogeneous, hard, and
brittle, is known as martensite (hard steel).

When the cooling' is carried out slowly, so that the mass passes
through a succession of equilibrium states, solidification takes place
at 1130°, with production of a heterogeneous mass of martensite
(2 per cent, of C) and scales of graphite. As the temperature falls
to 1000°, more graphite separates from the solid solution until the
martensite contains 1-8 per cent, of dissolved carbon. At this
point cementite, Fe3C, begins to separate. At 670°, a-ferrite begins
to separate, and the remaining solid solution then contains 0-9
per cent, of carbon. The solid solution will then slowly separate
at this temperature into a heterogeneous mixture of 87 per cent, of
soft a-ferrite and 13 per cent, of hard cementite, the mass, known
as pearlite, thus containing the three phases : a-ferrite + cementite
+ graphite. The addition of manganese, nickel, etc., retards the
conversion of y-ferrite into a-ferrite and /3-ferrite, and thus pro-
duces a more homogeneous steel, the Fe3C remaining in solution as
martensite.

These changes are attended with evolution of heat, which can be
followed by observing the temperature of the cooling metal at various
times by a pyrometer, and the separation of the various constituents
may be observed by quenching, polishing the steel, etching the surface
with reagents, and examining microscopically. The change taking
place at about 760° is the cause of recalescence, the sudden re -heating
of a mass of red-hot iron on cooling.

Wrought -iron is case-hardened by heating in contact with carbon
or potassium ferrocyanide, when a surface-layer of steel is produced.
Armour-plate is made by case -hardening a sheet of soft steel on one
side and then spraying it with cold water. It is pierced in a clean hole
by a soft-nosed shell, whereas hard steel splits in pieces. Nickel -chrome
steels form very tough armour-plates.

Pure iron. — The soft iron wire used for binding flowers contains
99-7 per cent, of Fe ; the perfectly pure metal is obtained by reducing
pure ferric oxide in hydrogen, or by electrolysis of a solution of
1 part of ferrous chloride. 1 part of calcium chloride, and 1 -6 parts
of water at 110°. It is a soft, almost white, metal, sp. gr. 7-86,
m.-pt. 1510°, b.-pt. 2450°. Iron is the most ductile and tenacious
of all metals except nickel and cobalt. It is permeated by hydrogen
at a red heat, and burns brilliantly in oxygen when heated to
redness. Powdered iron prepared by reduction at a fairly low tem-
perature is pyrophoric.

The rusting of iron. — Iron when exposed to ordinary moist air
is quickly corroded to a reddish-brown rust, consisting chiefly of

984 INORGANIC CHEMISTRY CHAP.

hydrated ferric oxide, 2Fe203.3H20. The conditions under which
rusting takes place have been investigated by several experimenters,
with divergent results. The homogeneity or otherwise of the metal
and its purity affect the results. The presence of water is essential,
and according to some experimenters the presence of carbon dioxide
is also necessary. Freshly-formed rust usually contains con-
siderable quantities of ferrous hydroxide and carbonate, indicating
that the formation of these compounds is probably the first step
in the corrosion of the metal.

Grace Calvert (1876) and Crum Brown (1888) suggested the
following reactions leading to the formation of rust :

1. Fe + H20 + CO2 = FeC03 + H2.

2. 4FeC03 + 6H2O + O2 = 4Fe(OH)3 -f 4CO2.

According to G. T. Moody (1906), pure iron does not rust in the
presence of water and air if every trace of carbon dioxide is excluded.
The iron first passes into solution, when carbon dioxide is present, as
ferrous bicarbonate. Fe(HC03)2, which then undergoes oxidation by
dissolved oxygen, with precipitation of ferric hydroxide, according
to the above equations. The addition of alkalies to the water, by
diminishing the content of carbonic acid, retards the rusting of iron.

EXPT. 336. — Take four lots, (a), (6), (c), (d), of clean iron nails.

(a) Boil ordinary tap-water in a test-tube until it begins to " bump,"
showing that dissolved air has been expelled. Drop the nails (a) into
the water, and boil again for half a minute. Pour melted vaseline over
the surface of the water. This excludes air, so that iron and water
alone are present.

(b) Place nails (b) in a test-tube full of ordinary water. In this
case iron, much water, and air are present.

(d) Place nails (d) in a desiccator over sulphuric acid. In this case
iron and air alone are present.

Leave* the four specimens for a few days, and examine the iron.
Rusting should have occurred only in cases (b) and (c).

EXPT. 337. — Pour 100 c.c. of 15 per cent, caustic potash solution into
a 500 c.c. flask, fitted with a cork partly bored, and shake. Allow
the flask to stand for two days. Boil a large bright nail with distilled
water, as described above, and push it through the cork into the flask,
leaving a short length outside. Allow to stand for a few days. The
part of the nail inside the flask, which is exposed to air and water in
the absence of carbon dioxide, does not rust, whilst the part outside*
exposed to moisture and air containing carbon dioxide, will rust.

EXPT. 338. — It will be noticed in Expt. 336 (6) that the undersides
of the nails remain bright, and rust is deposited on the top, exposed to

XLVIII IRON 985

air. This indicates that the iron passes into solution, and the solution
is then oxidised by the air. Place a number of bright nails in a jar,
cover them with a piece of hardened filter paper, and pour boiled dis-
tilled water into the jar. Rust is deposited above the filter-paper.

According to another theory of rusting, the different parts of a
piece of iron act as poles of voltaic cells and solution of the metal
occurs as the result of local action. This is quite compatible with
the fact that oxidation occurs only in solution, since ferrous ions may
be formed initially, but the action of carbonic acid is not intro-
duced. Lambert (1912), who took the most rigid precautions to
exclude carbon dioxide, found that, although homogeneous iron
does not rust even in ordinary air, ordinary iron rusts in the absence
of carbon dioxide.

EXPT. 339» — Prepare a solution of agar-agar in hot water (1£ per
cent.), and add a little sodium chloride and phenolphthalein. Pour
some of the solution over a clean plate of iron in a glass dish. The agar
sets to a jelly. After some hours red patches appear, indicating the
formation of caustic soda by electrolysis. If potassium ferricyanide
and phenolphthalein are added to the agar, and the hot solution is poured
over clean iron nails, the anodes become blue, from reaction of ferri-
cyanide with ferrous ions, and the cathodes red, from the caustic
potash formed.

The cause of rusting on this theory is electrolytic and due to the
different solution pressures (p. 884) of different parts of the metal.

Iron is protected from rusting by painting, or whitewashing with
lime. Pipes are also protected by heating and dipping into a solution
of coal-tar pitch in coal-tar naphtha, when an impervious coating is
formed (Angus Smith's compound). In the Barff process, the iron is
heated to redness and steam blown over it, when an adherent layer of
ferroso -ferric oxide is formed. This is used in treating cans for fruit,
etc., instead of tinning. The layer of oxide is removed by heating with
water containing magnesium chloride, which explains the corrosive
action of sea-water on boilers.

Passive iron. — Iron is rendered passive (p. 949) by immersion in
fuming nitric acid, chloric acid, chromic acid, or hydrogen peroxide,
or by making it the anode in electrolysis. The metal is then insoluble
in dilute acids, and does not precipitate copper from a solution of copper
sulphate (Kier, 1790). The passivity is removed by touching with
active iron under the surface of dilute sulphuric acid. The passivity may
be due to a film of oxide, Fe3O4 ; it is removed by heating in hydrogen.

Salts and ions of iron. — Iron readily dissolves in dilute hydro-
chloric or sulphuric acids, producing ferrous salts, the solutions of
which contain the bivalent ferrous ion, Fe" : Fe + 2H' =. Fe" + H2-

986 INORGANIC CHEMISTRY CHAP.

In dilute nitric acid, no gas is evolved, the acid is reduced, and
ammonium nitrate formed :

8Fe -f 20HN03 = 8Fe(NO3)2 + 2NH4N03 + 6H20.

Solutions containing the ferrous ion are nearly colourless, but usually
possess a green tinge, due apparently to the presence of traces of the
ferric ion, Fe'". They have an inky taste, and readily undergo
oxidation by atmospheric oxygen, insoluble basic ferric salts being
deposited.

Ordinary ink contains ferrous sulphate, tannin (an organic substance
obtained from gall-nuts or oak-bark), and gum. This solution, contain-
ing ferrous tannate, has a very pale colour, so that a solution of indigo-
sulphuric acid, or a blue dye is added. On exposure to air, the ferrous
tannate is oxidised to ferric tannate, which has an intense black colour,
and the blue colour of the ink gradually changes to deep black.

The ferrous ion is readily converted by oxidation into the ferric
ion, Fe'", which is also almost colourless, the red or brown colour
of ordinary solutions of ferric salts being due to the undissociated
compound, to basic compounds, or to colloidal ferric hydroxide
formed by hydrolysis. If these brown solutions are mixed with
concentrated nitric acid they become nearly colourless ; with con-
centrated hydrochloric acid they become deep yellow, the colour
of undissociated ferric chloride.

The oxidation of ferrous to ferric salts may be effected : (i) by
atmospheric oxygen in neutral solutions, when insoluble basic
ferric salts are precipitated ; (ii) by chlorine or bromine :
2Fe" -}- C12 = 2Fe*" + 2C1'. The reaction with iodine is rever-
sible : 2Fe" -f- I2^2Fe*" -f- 21' ; ferric chloride liberates iodine
from potassium iodide and iodine oxidises ferrous chloride to ferric
chloride. Ferrous salts are also oxidised by boiling with nitric
acid or aqua regia.

Ferric salts are reduced to ferrous salts by nascent hydrogen, in
acid solution, say by a mixture of zinc and hydrochloric acid :
2Fe'" + H2 = 2Fe" -f 2H\

A solution of silver nitrate oxidises a ferrous salt, silver being
precipitated. In this case a transfer of ionic charge occurs :
Fe" + Ag- = FeV" + Ag.

Oxides of iron. — Iron forms three well-defined oxides :
ii

(1) Ferrous oxide, FeO, a strong base, corresponding with the

ferrous salts, e.g., FeS04 (which may be written on the old dualistic
notation as FeO,S03) ; these are formed by dissolving iron in dilute
acids.

(2) "Ferric oxide, Fe203, a fairly strong base, corresponding with

XL vm IRON 987

III
the ferric salts, e.g., Fe2(S04)3 (Fe203,3SO3) ; with very strong bases

unstable ferrites, e.g., Na2Fe204, are formed, so that ferric oxide
has also feebly acidic properties.

II in

(3) Ferroso-ferric oxide, Fe3O4. probably ferrous ferrite, Fe(Fe02)2,
ii in

or FeO,Fe203.

(4) The unstable ferrates, e.g., K2FeO4; correspond with an
unknown acidic trioxide, Fe03.

FERROUS SALTS.

Ferrous chloride, FeCl2. — This salt is deposited from solutions
of iron in hydrochloric acid in bluish-green monoclinic crystals,
FeCl2,4H20, which oxidise slightly, and become green, in the air.
The anhydrous chloride, FeCl2, is obtained in white lustrous scales
on heating iron in hydrogen chloride : Fe + 2HC1 = FeCl2 + H2.

EXPT. 340. — Place a spiral of iron wire in a hard glass tube and pass
over it dry hydrogen chloride. Heat the spiral strongly, and notice
the sublimation of ferrous chloride. The escaping hydrogen may be
ignited.

Anhydrous ferrous chloride volatilises at a bright red heat ; its
vapour density indicates that molecules of Fe2Cl4 and FeCl2 are
present. The density becomes normal between 1300° and 1500° :
Fe2Cl4 ^ 2FeCl2. On heating the substance in air, oxidation
occurs ; ferric chloride volatilises, and ferric oxide remains :
12FeCl2 -f 302 = 2Fe203 -f 8FeCl3. When ferrous chloride is
heated in steam, hydrogen is evolved :

3FeCl2 + 4H20 = Fe304 + 6HC1 + H2.

Ferrous bromide, FeBr2, and ferrous iodide, FeI2, are prepared simi-
larly to the chloride, and form the crystalline hydrates FeBr2,6H2O
and FeI2,5H2O. They are also formed by adding the halogen to iron
filings (in excess) in presence of water. If excess of iodine is used,
ferroso-ferric iodide, Fe3I8, is formed, which gives with caustic potash
a black precipitate of ferroso-ferric hydroxide: Fe3I8 + 8KOH =
Fe3(OH)8 + SKI. This reaction is used in the preparation of potass-
him iodide (p. 792).

Ferrous sulphate, FeS04.— This is the most important ferrous salt,
and is obtained by dissolving iron in dilute sulphuric acid (p. 185),
or by the slow oxidation of marcasite, or " coal-brasses." FeS2,
by air in presence of water. The common form is green vitrtol,
FeS04,7H2O, crystallising in monoclinic crystals isomorphous with

988 INORGANIC CHEMISTRY CHAP.

Epsom salts, MgS04,7H20. If a crystal of white vitriol, ZnSO4,7H20,
is placed in the saturated solution rhombic crystals of FeS04,7H20,
isomorphous with that salt, are deposited, whilst blue vitriolt
CuS04,5H2O, induces the deposition of tridinic isomorphous crystals
of FeS04,5H20. By precipitating the solution with alcohol, or by
heating green vitriol in a vacuum at 140°, the monohydrate,
FeS04,H2O, is formed, and this on heating at 300° in absence of air
leaves the white, amorphous anhydrous salt, FeS04. Crystalline
hydrates with 6, 3, and 2H20 are also known.

Ferrous sulphate readily forms double-salts with the sulphates of
the alkali-metals, R2S04,FeSO4,6H20. If equimolecular amounts
of ferrous sulphate and ammonium sulphate are dissolved in separate
amounts of hot water, and the filtered solutions mixed, ferrous
ammonium sulphate, or Mohr's salt, (NH4)2S04,FeS04,6H2O, is
deposited on cooling in light bluish-green monoclinic crystals, which
may also be deposited in the form of a practically white powder on
adding alcohol to the solution (cf. FeS04,7H20). The crystals are
stable in the air and the solution is much less readily oxidised by
atmospheric oxygen than ferrous sulphate or chloride. Mohr's salt
is therefore used in volumetric analysis for standardising solutions of
potassium permanganate or dichromate ; it contains almost exactly
one -seventh of its weight of ferrous iron.

Ferrous carbonate, FeC03. — This compound occurs as siderite,
or spathic iron ore, in rhombohedra isomorphous with calcite. It is
formed as a white precipitate on addition of an alkali carbonate to
ferrous salts. The precipitate rapidly becomes green, and finally
brown, on exposure to air, owing to oxidation to ferric hydroxide.
The addition of sugar retards the oxidation. Ferrous carbonate
dissolves in water containing carbonic acid, forming ferrous bicar-
bonate, Fe(HC03)2, which is sometimes present in rivers. On
exposure to air, red ferric hydroxide is precipitated :

2Fe(HC03)8 + 0 = Fe203 + 4C02 -f 2H20.

Plants absorb iron from the soil as the bicarbonate.

Ferrous hydroxide, Fe(OH)2. — This compound is thrown down as
a white precipitate when caustic soda is added to a pure solution of
a ferrous salt, with absolute exclusion of air. It is insoluble in excess
of alkali, unless the latter is very concentrated, but dissolves slightly
in ammonium salts. The precipitate rapidly becomes green in the
air, from formation of Fe3(OH)8, and finally brown, forming
Fe(OH)3.

To obtain the original solution free from ferric salts, it is warmed
with a little iron and dilute acid in a flask fitted with a tube dipping
under water.

Ferrous oxide, FeO, is formed as a pyrophoric black powder by
reducing ferric oxide with hydrogen at 300°, or by adding ferrous

XLVIIT IRON 989

oxalate (obtained by precipitating ferrous sulphate with ammonium
oxalate) to boiling caustic potash. It melts at 1420°. A mixture
of FeO and finely-divided iron, which is pyrophoric, is obtained by
heating ferrous oxalate at 150-160° in absence of air : FeC2O4 =
FeO -f CO + C02. Ferrous oxide is reduced to metallic iron by
hydrogen at 700-800°.

FERRIC SALTS.

Ferric hydroxide, Fe(OH)3. — If ammonium chloride and ammonia
are added to a solution of a ferric salt, such as is obtained by oxidising
ferrous sulphate with nitric acid or aqua regia, a reddish -brown,
gelatinous precipitate of ferric hydroxide, Fe(OH)3, is thrown down,
which is slimy in the cold, but becomes flocculent on boiling. It is
practically insoluble in water and alkalies, and is the form in which
iron is separated in quantitative analysis. On prolonged boiling in con-
tact with the solution, it becomes sparingly soluble in acids, whereas
the freshly-formed precipitate is readily soluble. This appears
to be due to loss of water. The precipitate is colloidal, and dries
to a gum-like mass of indefinite composition. Crystalline hydrated
ferric oxides occur in the minerals limonite, 2Fe203,3H20, gothite,
Fe203,H20, and hydrohcematite, 2Fe203,H20. On ignition, brownish-
red ferric oxide, Fe203, is formed, which in this state is nearly
insoluble in acids ; it dissolves in concentrated hydrochloric acid
only after digestion for several days, more easily in presence of
ferrous salts. The best solvent is a boiling mixture of 8 parts of
H2SO4 and 3 parts of water. If a current of hydrogen chloride is
passed over the strongly -heated oxide, the latter becomes crystalline.
Ferric oxide melts at 1563°. Red varieties of ferric oxide are formed
by igniting ferrous sulphate in the air, and are used as paints or
as a polishing powder (rouge, crocus, colcothar).

Colloidal ferric hydroxide is obtained by dissolving freshly -precipi-
tated ferric hydroxide in a concentrated solution of ferric chloride, and
dialysing. The blood-red solution (dialysed iron) is a positive colloid
(p. 888), and is readily precipitated by salts. On adding concentrated
hydrochloric acid, the solution is slowly converted into yellow ferric
chloride. If glycerin, sugar, tartaric acid, etc., are added to a solution
of a ferric salt, the latter is not precipitated by ammonia, but a clear
brown colloidal solution is formed. If organic matter is present in a
substance, it must therefore be destroyed by ignition before the ordinary
group -reagents of qualitative analysis are used.

If ferric oxide is strongly heated with sodium carbonate, sodium
ferrite, Na2Fe204 (Na20,Fe2O3) is formed : Na.2C03 + Fe2O3 =
Na2Fe204 -f C02. On treating the mass with hot water, the ferrite

990 INORGANIC CHEMISTRY CHAP.

is decomposed and a solution of caustic soda is produced, the
ferric oxide being regenerated : Na2Fe204 + H2O = 2NaOH -f-
Fe2O3. This is the Lowig process for the manufacture of caustic
soda.

Ferroso-ferrie oxide, Fe304. — This oxide is strongly magnetic
and is formed by heating iron to redness in air (" smithy scales "),
or in steam. The pure oxide is obtained as a black powder by
reducing Fe2O3 at 400° in a current of hydrogen and steam. It
melts at 1540°, and is cast into electrodes, since it resists acids and
chlorine when fused. Ferroso-ferric hydroxide, Fe3(OH)8, is formed
as a black precipitate by adding caustic soda to a mixture of a
ferrous and a ferric salt. It dissolves in hydrochloric acid to a green
solution, from which crystals of ferroso-ferric chloride, Fe3Cl8,18H20,
separate on evaporation. The oxide Fe304 appears to be ferrous
ferrite, Fe:(Fe02)2.

Ferric chloride, FeCl3. — This is the most important ferric salt.
It is obtained anhydrous, in iron-black crystals with a green irides-
cence, on heating iron in chlorine (cf. Expt. 328). These volatilise
on heating, and at 444° the vapour density corresponds with Fe2Cl6.
With rise of temperature the vapour density falls, owing to disso-
ciation, and at 750° becomes nearly equal to that required by the
formula FeCl3, although it still decreases, probably owing to disso-
ciation into FeCl2 and chlorine : Fe2Cl6 ^± 2FeCl3 ; 2FeCl3 ^±
2FeCl2 + C12.

Temperature 448° 518° 606° 750° 1050° ' 1300°

A (H = 1) . . 151 138 121 78 76-3 734

In solutions in alcohol and ether the molecular weight of ferric
chloride corresponds with FeCl3. The anhydrous chloride is also
soluble in benzene. These solutions exhibit the bright yellow
colour of FeCl3 molecules. Aqueous solutions containing excess of
hydrochloric acid are also bright yellow. In alcoholic solution,
containing water, ferric chloride is reduced on exposure to light,
and green crystals of FeCl2,2H20 are deposited.

Aqueous solutions of ferric chloride are produced by dissolving
ferric hydroxide in hydrochloric acid, or by saturating solutions of
ferrous chloride with chlorine. On evaporation, crystals containing
Fe2Cl6,12H20, Fe2Cl6,7H20, Fe2Cl6,5H20, FeaCl6,2H20, and Fe2Cl6
are deposited at 37°. 32-5°, 56°, 73-5°, and (from solutions con-
taining more ferric chloride than corresponds with Fe2Cl6,2H20) at
60°. respectively. If the solution is evaporated to the composition
FeCl3,6H2O, yellow crystals are deposited on cooling, which are
readily soluble in water. Ferric chloride solution is used as a
styptic, i.e., in stopping bleeding. It coagulates the blood, forming
a clot. The solution is strongly acid, due to hydrolysis (p. 989) :
FeCl3 + 3H20 ^ Fe(OH)3 + 3HC1. On heating the hydrated salts,

XL vni IRON 991

hydrochloric acid is evolved, and a basic salt, or finally ferric oxide,
is left.

Garnet-red double salts are formed from ferric chloride and other
chlorides : FeCls,2KCl,HaO, FeCl3,2NH4Cl,H2O, FeCl3,MgCl2,H2O.

Ferric fluoride, FeF3, is a white, difficultly soluble salt, only slightly
ionised in solution. It forms double fluorides, e.g., Na3FeF6, analogous
to cryolite (p. 898). The bromide. FeBr3, is formed similarly to the
chloride, but the iodide does not appear to exist (cf. p. 986).

Ferric phosphate, FeP04,2H20, is obtained as a white precipitate,
insoluble in acetic acid, but soluble in mineral acids, when sodium
phosphate is added to a ferric salt. It is used in the separation
of phosphates in qualitative analysis (p. 630).

Ferric sulphate, Fe2(S04)3. — A solution of this salt is obtained by
boiling ferrous sulphate with sulphuric and nitric acids. Nearly
pure nitric oxide is evolved : 6FeS04 + 3H2S04 + 2HN03 =
3Fe2(S04)2 -f 2NO + 4H20. A black solution containing FeSO4 -NO
(p. 580) is first formed. A similar reaction occurs with ferrous
chloride and aqua regia, ferric chloride being produced. The reaction
is used in the estimation of nitrates (Schloesing) ; the nitric oxide
evolved is measured. Prolonged boiling, preferably under reduced
pressure, is, however, necessary to complete the reaction.

Ferric sulphate is also formed by evaporating ferrous sulphate with
concentrated sulphuric acid :

2FeS04 + 2H2S04 = Fe2(S04)3 + S02 -f 2H20.

Anhydrous ferric sulphate is a yellowish-white powder, dissolving
only very slowly in water, but ultimately forming a very con-
centrated solution. This is slightly yellow owing to hydrolysis,
but becomes nearly colourless on addition of sulphuric acid. With
potassium and ammonium sulphates ferric sulphate forms iron

alums, e.g., (NH4)2S04,Fe2(S04)3,24H20, with a slight yellow tinge
when pure, but often violet, possibly owing to the presence of man-
ganese. These are readily soluble in water, and are not appreciablv
hydrolysed. The potassium alum; K2S04,Fe2(S04)3,24H2O, does
not crystallise so readily as the ammonium salt.

On heating ferric sulphate, sulphur trioxide is evolved, the
reaction being reversible : Fe2(S04)3 ^± Fe203 -f 3S03.

Ferric nitrate is obtained by dissolving iron in fairly concentrated
nitric acid ; the dark brown solution (used as a mordant) deposits
colourless cubic [Fe(NO3)3,6H2O] or monoclinic [Fe(NO3)3,9H2O] crystals.

If iron is dissolved in sulphurous acid no gas is evolved. The solution
deposits colourless crystals of ferrous sulphite, and a solution of ferrous
thiosulphate is left : 2Fe + 3H2SO3 = FeSO3 + FeS2Os + 3H2O.

Iron salts act as catalysts in many reactions. Thus, if hydrogen

992 INORGANIC CHEMISTRY CHAP.

peroxide is added to potassium iodide and starch acidified with acetic
acid, iodine is only slowly liberated, but on addition of a drop of ferrous
sulphate the reaction is instantaneous. The iron in chlorophyll and
haemoglobin may have something to do with the activity of these
substances.

Iron earbonyls. — When carbon monoxide is passed over finely
divided iron at 120°, iron pentacarbonyl, Fe(CO)5, is produced. It is
a pale yellow, viscous liquid, b.-pt. 102-5°, fr.-pt. —20°. The vapour
is decomposed on passage through a tube heated to 180°, a mirror
of metallic iron being deposited. The vapour density at 129°, and
the freezing point of the solution in benzene, correspond with the
above formula. Iron pentacarbonvl is decomposed by air and mois-
ture, and by acids : Fe(CO)5 -f H2SO4 = FeS04 -f- 5CO + H2. On
exposure to light, diierro-nonacarbonyl is formed, the reaction being
reversed in darkness : 2Fe(CO)5 =± Fe2(CO)9 + CO. Fe2(CO)9 forms
orange crystals, decomposing on heating : Fe2(CO)9 = Fe(CO)5 +
Fe + 4CO. If a solution of Fe2(CO)9 in toluene is heated to 50° it
becomes intensely green, and green crystals are deposited, which are
a polymerised form of iron tetracarbonyl, Fe(CO)4.

Iron pentacarbonyl is formed in traces when water-gas (p. 705) is
passed through iron pipes. Such gas deposits Fe2O3 on incandescent
mantles in gas-burners.

Sulphides of iron. — Ferrous sulphide, FeS, is formed as a black
mass by heating iron filings together with sulphur, a considerable
amount of heat being evolved. It may be prepared by dipping a
white-hot bar of wrought-iron into molten sulphur in a crucible.
(Cast-iron is not attacked.) A mixture of iron filings and sulphur
when moistened becomes heated and forms FeS. Ferrous sulphide
in the pure state is a yellowish, crystalline mass with a metallic
lustre, melting at 1300°. The commercial substance is black or
dark-grey, and contains free iron. It dissolves readily in dilute acids,
and is used in the preparation of sulphuretted hydrogen (p. 483).
A greenish-black precipitate of ferrous sulphide is formed when
ammonium sulphide is added to a ferrous salt : (NH4)2S + FeS04 =
FeS -f(NH4)2S04. The precipitate dissolves slightly in excess of
the reagent when the latter contains polysulphides, forming a dark
greenish - black solution, probably containing a ferri - sulphide,
(NH4)FeS2, or (NH4)2S,Fe2S3.

Potassium ferri-sulphide, KFeS2, is formed in purple crystals by
fusing together iron, sulphur, and potassium carbonate, and extracting
with water. On heating in hydrogen it forms a f errosulphide, K2Fe2S3, or
K2S,2FeS. The sodium salt, NaFeS2.4H2O, occurs in the crude black
liquors obtained by lixiviating black-ash (p. 778). It is removed? and the

XLVIII IRON 993

soda liquor decolorised, by heating with zinc oxide, when Fe2O3 is pre-
cipitated, and white ZnS formed.

Iron sesquisulphide, Fe2S3, is formed as a yellow mass with metallic
lustre, by heating FeS with sulphur, or by heating iron powder in
H2S at 100° ; it is thrown down as a black precipitate by the action
of excess of ammonia and ammonium sulphide on a solution of a ferric
salt ; with excess of ferric salt a mixture of 2FeS and S is formed.
The mineral magnetic pyrites consists of compounds of FeS and Fe2S3
varying from 5FeS,Fe2S3 to 6FeS,Fe2S3. Tetraferric trisulphide,
Fe4S3, is said to be formed by heating iron in carbon disulphide vapour.

Iron disulphide, FeS2, occurs as iron pyrites and marcasite. Pyrites
(sp. gr. 5-19) is stable in air, marcasite (sp. gr. 4-68-4-85) oxidises in
moist air to ferrous sulphate. Pyrites crystallises in the regular
system, often in cubes, either plain or striated ; sixty -nine forms
have been described. It has a brassy-yellow colour (" fools'
gold "), is very hard, striking sparks from steel, and is not magnetic.
Marcasite occurs in rhombic crystals, usually in the form of radiating
nodules, and is white like tin. Pyrites often occurs in coal and is the
main source of the sulphur dioxide formed on its combustion. It is
found in masses having the form of wood, roots, etc., and has pro-
bably been formed by the reduction of solutions of ferrous sulphate
by organic matter. Pyrites is insoluble in dilute acids but dissolves
readily in concentrated nitric acid, with separation of sulphur, or
in aqua regia.

Ferric acid. — A mixture of one part of iron filings and two parts of
nitre deflagrates on heating, and the cold product dissolves in water
to form a purple solution (Stahl, 1702). This contains the potassium
salt of ferric acid, H2Fe04 (Fremy, 1841). The purple solution is
also produced by the electrolysis of caustic potash with a cast-iron
anode, or by passing chlorine through ferric hydroxide suspended in
potash. If excess of caustic potash is added, reddish-brown
potassium ferrate, K2Fe04, is deposited. On boiling, a yellow solu-
tion of potassium f write, K2Fe204, is produced, which rapidly de-
posits ferric hydroxide. On addition of barium chloride to the red
potassium ferrate solution, fairly stable barium ferrate, BaFeO4,H20,
is formed as a red precipitate.

Potassium ferrocyanide, K4Fe(CN)6. — No simple cyanides of iron
are known ; if potassium cyanide is added to a solution of ferric
chloride, cyanogen is evolved, and ferric hydroxide is precipitated :
2FeCl3 + 6KCN + 6H20 = 2Fe(OH)3 -f 6HCN + 6KC1. Many
complex cyanides, however,, are known. When nitrogenous organic
matter, such as horn or leather- clippings, is fused with potassium
carbonate and iron filings and the mass digested with water, the
solution deposits on evaporation yellow crystals of potassium
ferrocyanide, or yellow prussiate of potash, K4Fe(CN)6,3H2O. The

3 s

994 INORGANIC CHEMISTRY CHAP.

addition of a ferric salt to the solution gives a deep blue precipitate
of Prussian blue, the first ferrocyanogen compound to be discovered
(Diesbach, 1704). Macquer (1752) showed that potassium ferro-
cyanide was formed on boiling Prussian blue with potash, and Porret
(1814) observed that the former salt contained a peculiar acid,
ferrocyanic acid, H4Fe(CN)6, formed as a white precipitate on adding
an acid, and then ether, to a solution of the ferrocyanide. The
precipitate contains combined ether. Berzelius pointed out that
the yellow prussiate might be regarded as a double cyanide of
potassium and iron, 4KCN,Fe(CN)2, but since it shows none of the
reactions of iron or of cyanides, it is more properly regarded as the
potassium salt of the complex ferrocyanic acid : K4[Fe(CN)G].
One of the CN groups may be replaced by CO, H2O, NO, N0?, etc.

Potassium ferrocyanide is often prepared from the spent-oxide of
gas works (p. 682) by boiling with potash and crystallising. The
nitrogen of the coal is partly evolved as cyanogen, which collects in
the oxide purifiers in the form of Prussian blue. The salt is also
formed by adding excess of potassium cyanide to a solution of ferrous
sulphate, until the brown precipitate redissolves. The crystals are
yellow, tetragonal pyramids, which are unchanged in air but on
heating fall to a white powder of anhydrous salt. Potassium ferro-
cyanide is not poisonous. The sodium salt, Na4Fe(CN)6,10H20,
is prepared in a similar manner. Silver nitrate gives a white preci-
pitate of silver ferrocyanide, Ag4Fe(CN)6, with soluble f err o cyanides.

Potassium ferrieyanide, K3Fe(CN)6. — If chlorine is passed through
a solution of potassium ferrocyanide, the quadrivalent ferrocyanide

ion, Fe(CN)4"", is oxidised (p. 255) to the tervalent ferricyanide ion,
in

Fe(CN)6///,the two ions containing bi- andter-valent iron respectively:
2Fe(CN)6"" -f C12 = 2Fe(CN)6/// + 2C1'. At the same time a molecule
of chlorine gas is reduced to two chloride ions. The two salts
KC1 and K3Fe(CN)6 separate on evaporation from the yellowish-
brown solution, but by repeated recrystallisation potassium ferri-
cyanide, K3Fe(CN)6, is obtained in the pure state in the form of anhy-
drous dark-red monoclinic prisms (red prussiate of potash, L. Gmelin,
1822). It is an oxidising agent, converting litharge into lead
dioxide, and chromium sesquioxide into potassium chromate, when
these are boiled with the alkaline solution : 6KoFe(CN)6 -{- Cr203 -f-
10KOH = 6K4Fe(CN)6 + 2K2Cr04 + 5H2O (or 4Fe(CN)6"'2+Cr" +
10OH' = 4Fe(CN)6"" -f 2Cr04" + 5H20). It is used in organic chem-
istry for effecting oxidations. The solution is reduced by sodium
amalgam and by glucose in alkaline solution to ferrocyanide. The
alkaline solution is reduced by hydrogen peroxide, whereas an
acid solution of ferrocyanide is oxidised by the same reagent (p. 340).
Sodium ferricyanide, 2Na3Fe(CN)6,H2O, is obtained from sodium
ferrocyanide and chlorine.

XLVTII IRON 995

Ferricyanic acid, H3Fe(CN)6, is formed in brown needles by decom-
posing lead ferricyanide [obtained in brown crystals, Pb3(FeCy6)2,
16H2O, by mixing hot solutions of Pb(N03)2 and K3Fe(CN)e] with
sulphuric acid and evaporating. Silver salts give a red precipitate
of silver ferricyanide, Ag3FeCy6, with ferricyanides.

Prussian blue. — When a solution of ferrous sulphate is added to
a cold neutral solution of potassium ferrocyanide, a white precipitate

H ii

of potassium ferrous ferrocyanide, K2Fe(FeCy6), is formed, which
rapidly oxidises in air to /^-soluble Prussian blue, or potassium ferric

in ii

ferrocyanide, FeK(FeCy6),H20, insoluble in oxalic acid but soluble
in water. But if ferrous sulphate is added to an acid solution of
ferrocyanide the white precipitate formed, although similar to
the above, is less readily oxidised, and on exposure to air forms
y-soluble Prussian blue, probably of the same formula as the /3-blue,
but more stable to alkalies, acids, and ferric chloride.

When potassium ferrocyanide is boiled with dilute sulphuric
acid, hydrocyanic acid (p. 717) is evolved, and may be obtained in
the form of a dilute solution by attaching a Liebig's condenser to the

flask : 2K4FeCy6 + 3H2S04 = 3K2S04 + K2Fe(FeCy6) + 6HCN.

The pale yellow powder, K2Fe(FeCy6), left in the flask is much less
easily oxidised than the other two forms just described, but nitric
acid or hydrogen peroxide converts it into Williamson's violet,

in ii
KFe(FeCy6),H20.

By heating a solution of ferrocyanic acid at 110-120° in a sealed
tube a precipitate of the acid corresponding with the white precipi-

ii ii
tates, ferrous hydrogen ferrocyanide, H2Fe(FeCy6), is formed, which

on oxidation gives a violet compound, possibly HFe(FeCy6).

When a solution of potassium ferrocyanide is precipitated with
rather less than the equivalent of ferric chloride, and the precipitate
washed by decantation with potassium chloride solution, it forms

m H

a-soluble* Prussian blue, or a-ferric ferrocyanide, 4FeK(FeCy6),7H20.
This, when dried, has a bronze lustre and forms a beautiful deep
blue powder. It dissolves in water, forming a blue colloidal solution,
and is soluble in oxalic acid.

in H

Prussian blue may, of course, have the possible formulas : FeK(FeCy6)
II III
or FeK(FeCy6). K. A. Hofmann (1904) showed that hydrogen peroxide

in acid solution, which reduces ferricyanides to ferrocyanides, oxidises
ferrous ferrocyanides to Prussian blue ; the latter must therefore con-
tain the ferric iron in the basic radical. The same Prussian blue is

3 s 2

996 INORGANIC CHEMISTRY CHAP.

formed by precipitating a f erricyanide with a ferrous salt ; in this case
isomeric change must have occurred.

With excess of ferric chloride, the precipitate becomes insoluble
in water, and is called insoluble Prussian blue ; it has the formula
in ii

Fe4(FeCy6)3, but contains water which cannot be driven off by heat.
It is the ordinary Prussian blue of commerce.

The precipitate obtained by adding an excess of ferrous salt to
potassium f erricyanide, known as TurnbulPs blue, was formerly

ii in

considered to be ferrous ferricyanide, Fe3(FeCy6)2 ; it is, however,
identical with insoluble Prussian blue. A ferric salt with potassium
ferricyanide gives a deep brown solution, probably containing ferric
ferricyanide, but no precipitate. A little stannous chloride, or
granulated zinc and acid, added to the solution, precipitates Prussian
blue. If chlorine is passed through a boiling solution of potassium
ferrocyanide in the dark, a green precipitate of ferric ferricyanide,

ni m"
probably polymerised, Fe(FeCy6), is formed.

Sodium nitroprusside. — When potassium ferrocyanide is warmed
with 50 per cent, nitric acid, a brown solution is produced. When
the reaction has proceeded to such a stage that a slate-coloured
precipitate is formed with ferrous sulphate the liquid is cooled,
separated from the crystals of potassium nitrate, and neutralised
with sodium carbonate. The filtered solution on evaporation gives
red crystals, which may be freed from nitrate by repeated crystallisa-
tion, and consist of sodium nitroprusside, or sodium nitrosoferri-

iii
cyanide, Na2Fe(NO)Cy6,2H2O. It is used as a reagent, giving an

intense purple colour, due to the formation of Na3Fe(0:N-SNa)Cy5,
with alkali-sulphides, but not with free sulphuretted hydrogen.
With silver nitrate a solution of a nitroprusside (which soon
decomposes, and is made as required) gives a flesh-coloured
precipitate of the silver salt. By decomposing this with hydrochloric
acid, unstable free nitrosoferricyanic acid, H2Fe(NO)Cy5? is formed.
This is also formed by passing nitric oxide into an acidified solution
of potassium ferrocyanide.

Ferric thiocyanate, Fe(CNS)3. — This salt is formed when potassium
or ammonium thiocyanate is added to a solution of a ferric salt. It
has a deep blood-red colour, and its formation is a- delicate test for
the ferric ion. The reaction is reversible (p. 350) : FeCl3 + 3KCNS;=i
Fe(CNS)3 + 3KC1. If the solution is shaken with ether, the latter
dissolves the ferric thiocyanate, leaving the aqueous layer colourless.
Mercuric chloride discharges the red colour of the aqueous solution ;
the mercury salt, which is only slightly ionised, is formed from the
ferric salt. Reducing agents form ferrous thiocyanate, also colour-

XLVIII IRON 997

less in solution. The red colour of the ferric salt is due to the un-
dissociated molecules.

Peculiar series of complex iron compounds, containing nitrogen and
sulphur, are known. If a solution of ferrous sulphate in excess of a thio-
sulphate is saturated with nitric oxide, crystalline iron dinitrosothiosul-
phates are formed, e.g., reddish-brown leaflets of K[Fe(NO)2S2O3],H2O ;
brilliant jet-black crystals of Rb[Fe(NO)2S2O3],H2O. If nitric oxide
is passed through a suspension of precipitated ferrous sulphide in
dilute solutions of sulphides, black compounds (Roussin's salts) are
formed, e.g., KFe4(NO)7S3, which form dark brown solutions with
water. By the action of alkalies on these, salts such as K2Fe2(NO)4S2
are formed.

EXERCISES ON CHAPTER XLVIII

1. Discuss the position of the transitional elements in the Periodic
System. What other elements show analogies to them ?

2. What are the important ores of iron ? How is cast-iron obtained
from the ores ?

3. How are steel and malleable iron produced from cast-iron ? What
is the cause of the different properties of these varieties of iron ?

4. Describe briefly the changes occurring in the hardening and tem-
pering of steel. What special varieties of steel are made ?

5. How are the oxides of iron prepared ? Discuss their properties
with special reference to their acidic and basic character.

0. How are ferrous and ferric chlorides prepared ? In what way can
ferrous chloride be converted into ferric chloride, and vice versa ?

7. Describe the preparation and properties of ferrous and ferric
sulphates. What important double salts of these compounds exist ?
How would you prepare them from metallic iron ?

8. How are iron carbonyls prepared ?

9. Describe the preparation of (a) potassium ferrocyanide, (6)
potassium ferricyanide, (c) sodium nitroprusside.

10. Give a brief account of the cyanogen compounds of iron.

CHAPTER XLIX

COBALT AND NICKEL

COBALT. Co = 58-50.

Cobalt. — The copper-miners of the Hartz Mountains frequently
obtained ores looking like copper-ore ; these gave an unpleasant
smell of garlic on roasting, and furnished no copper. The miners
attributed their occurrence to the pranks of an evil spirit, kobold,
and the material was called " false-ore," or cobalt.

In the mines of Chemnitz, the Baroness de Beausoleil claimed to have
seen " aged dwarfs, three or four hands in height, clothed like miners in
an old robe, with a leather apron hanging from the waist, a white coat,
and a cowl. They carried lamps and picks, and were terrifying appari-
tions to those who had not the assurance gained by experience in the
mines."

The residue left after roasting cobalt, called zaffre (impure cobalt
arsenate), was found to give on fusion with sand and potassium
carbonate a beautiful blue glass, called smalt. The despised ore
began to be valued, and the work of the " evil spirit of the mine "
beautified the magnificent stained glass windows of the churches.
The blue colour, believed to be due to arsenic, was shown by Brandt
(1735) to originate from a new metal contained in the ore, which he
called cobalt rex : Bergman (1780) investigated its properties, and the
metal then became known simply as cobalt. The " false-ore " is
an arsenide of cobalt, iron, and nickel, (Fe,Ni,Co)As2 (in the pure
state, CoAs2), it is known as speiss cobalt, or smaUite. A similar
ore is linnceite, (Co,Ni,Fe)3S4. Cobalt is also found as cobalt glance,
or cobaltile, (Co,Fe)SAs, and as cobalt bloom, Co3(AsO4)2,8H20, but is
now mainly obtained from the silver ores of Cobalt City, Ontario, and
of New Caledonia, where arsenides of nickel and cobalt occur
plentifully.

Metallurgy of cobalt. — The ore is roasted to free it from arsenic
and sulphur, and fused in a blast-furnace with limestone and sand as
a flux. The iron passes into the slag, and impure oxide of cobalt
(speiss) settles out. This is extracted with hydrochloric acid, metals
of the second group precipitated with sulphuretted hydrogen, and
from the filtrate, boiled with bleaching powder to oxidise the iron,
the latter is thrown down by milk of lime. The nitrate contains

CH. XLTX COBALT AND NICKEL 999

cobalt and nickel salts, and excess of lime ; if a further addition of
bleaching powder is made the cobalt is precipitated as a black
peroxide, Co203 (p. 162), which is reduced by heating in hydrogen.
Nickel is precipitated from the nitrate by adding excess of milk of
lime. The Canadian ores are roasted, or leached with ferrous sul-
phate solution, when sulphates are formed. These are soluble and the
cobalt is precipitated as before. The metal may also be precipitated
by potassium nitrite as potassium cobaltinitrite, K3Co(NO2)6. The
metal is prepared by the electrolysis of a solution of the sulphate,
CoS04, containing ammonium sulphate and ammonia,. It is tena-
cious, silver-white in colour, with a slight pink tinge, readily polished,
and shows a high lustre. Its specific gravity is 8-8, it is magnetic
up to 1150°, and melts at 1530°. Cobalt slowly oxidises on heating
in air. It absorbs 59-153 volumes of hydrogen when in a finely-
divided state. The metal dissolves slowly in dilute sulphuric and
hydrochloric acids, and readily in nitric acid. It can become
passive in nitric acid.

Cobalt oxides. — The solution of cobalt in nitric acid contains
cobalt nitrate, Co(N03)2,6H20, which may be obtained on evaporation
in pink crystals, which lose water over sulphuric acid to form a pink
powder. This readily decomposes on heating, leaving black
cobalto-cobaltic oxide, Co3O4. The solutions of cobalt salts are pink,
and contain the bivalent cobalt ion, Co". On addition of caustic
potash, a bluish -violet precipitate of a basic salt is thrown down,
which on boiling is converted into pink cobaltous hydroxide, Co(OH)2.
When heated out of contact with air, this forms an olive-green powder
of the basic cobaltous oxide, CoO. Cobaltous hydroxide dissolves only
in traces in excess of caustic potash, but is readily soluble in ammonia,
a yellowish-brown solution of a complex compound being formed.
This deposits cobaltous hydroxide on dilution, but readily absorbs
oxygen from the air to form stable complex compounds known as
cobalt ammines (p. 1001).

On gently igniting cobalt nitrate, a sesquioxide, Co203, is obtained
as a dark brown powder. This is probably the oxide formed when
bleaching-powder; or iodine and caustic potash, is added to a solu-
tion of a cobaltous salt, although this may be CoO2.

When hydrogen peroxide is added to cobaltous hydroxide suspended
in water, the filtrate is acid, and becomes green on addition of potassium
hydrogen carbonate. Cobaltic acid, H2CoO3, or a complex cobalt com-

ITI
pound, [Co(KCO3)2]2O, may be formed.

Both CoO and Co203 on ignition in air form Co304, and when
heated in hydrogen all the oxides are reduced to the metal.

A solution of cobalt nitrate is used in blowpipe analysis for the
detection of zinc and aluminium compounds. The ignited residue on

1000 INORGANIC CHEMISTRY CHAP.

charcoal is moistened with one drop of dilute cobalt nitrate and reheated.
Zinc gives a green mass (Rinman's green, cobalt zincate, CoZnO2) ;
aluminium a blue mass [Thenard's blue, cobalt aluminate, Co(AlO2)2],
although blue masses are also produced with phosphates. Magnesia
gives a pink mass. Cobalt salts give a beautiful dark blue borax bead.

Cobalt salts. — With the exception of the blue cobaltie sulphate,
Co2(SO)3,18H2O, which forms alums, and is obtained by the electro-
lytic oxidation of cobaltous sulphate ; and the dark brown unstable
solutions of Co2O3 in acids, which contain the tervalent cobaltic ion,
Co"", all the simple cobalt salts are derived from bivalent cobalt.
The complex cobaltaminines contain tervalent cobalt.

Cobaltous chloride, CoCl2,6H2O, is obtained in dark-red, deliques-
cent crystals from a solution of cobalt or the oxide in hydrochloric
acid. It forms a number of lower hydrates. The anhydrous salt
(and the lowest hydrates), obtained by heating, are deep blue in
colour. A solution of cobalt chloride is used as a sympathetic ink ;
the writing is almost invisible, but becomes blue on holding the paper
before the fire. On standing in moist air, the colour again disappears.

Other sympathetic inks, which are " irreversible," are dilute sul-
phuric acid, which chars the paper on heating, and a lead or bismuth
salt, which becomes black on exposure to sulphuretted hydrogen. The
latter is the original invisible ink (N. Lemery, 1681). The cobalt ink
was introduced in 1705.

The pink solutions of cobalt chloride also become blue on heating,
or addition of hydrochloric acid, or alcohol. A complex blue anion,
CoCl4", appears to be formed: 2CoCL2^Co" -f- CoCl4", which
moves to the anode in electrolysis.

Cobaltous sulphate, CoS04,7H20, is isomorphous with the vitriols
(e.g., FeSO4,7H20). It crystallises with different amounts of water,
according to the temperature ; the solution at 40-50° deposits
CoS04,6H2O, and when poured into concentrated sulphuric acid,
CoS04,4H2O. The anhydrous salt is pink. Double sulphates,
e.g., K2S04?CoS04,6H20, are known.

Cobalt sulphide, CoS, is precipitated by ammonium sulphide, or
by sulphuretted hydrogen in presence of sodium acetate. It is
black, and, although not precipitated by sulphuretted hydrogen
from acid solutions, it is insoluble in dilute acids ; it is soluble in
aqua regia.

By heating CoS with sulphur in a current of hydrogen, CoS2, Co2S3,
and Co2S are said to be formed. A persulphide, possibly Co2S7, is
formed as a black precipitate with yellow ammonium sulphide. Cobalt-
ous carbonate, CoCO3, forms'a bright red precipitate. Cobalt carbonyl,
Co2(CO)8, is obtained in orange-red crystals, m.-pt. 51°, by heating
cobalt at 150° in carbon monoxide under 30 atm. pressure. At 60°, it

XLIX COBALT AND NICKEL 1001

forms Co(CO)3, giving black crystals from a solution in benzene. The
carbide, Co3C, formed at high temperatures, is almost completely decom-
posed on cooling.

Blue cobalt glass, and the blue glazes on porcelain contain the
orthosilicate, Co2SiO4. If stannic oxide is added, the orthostannate,
Co9SnO4, is formed.

Complex cobalt compounds. — When potassium cyanide is added to
a solution of a cobalt salt, a brownish -white precipitate of cobaltous
cyanide, Co(CN)2, is thrown down. This dissolves in excess of cyanide,

forming potassium cobaltocyanide, K4Co(CN)6, analogous to the
ferrocyanide, which is thrown down as a deep amethyst-coloured
powder by alcohol. If a little acetic or hydrochloric acid is added
to the solution, and the latter boiled in an evaporating dish for a few

minutes, oxidation occurs and potassium cobalticyanide, K3Co(CN)6,
analogous to the ferricyanide, is formed : 2K4Co(CN)6 -f- H2O + O
= 2K3Co(CN)6 -f 2KOH. An equivalent amount of hydrogen
peroxide is contained in solution, so that autoxidation probably
occurs : H2O -f- 02 = H202 + 0. The cobalticyanide forms stable,
yellow crystals, isomorphous with K3Fe(CN)6. It gives a blue
precipitate with copper sulphate, Cu3(CoCy6)2, and a white precipitate
with silver nitrate, from which crystalline cobalticyanic acid, H3CoCy6,
is formed with H2S. Cobalticyanides give none of the reactions of
cyanides or of cobalt, and are not decomposed by concentrated
nitric acid.

Potassium nitrite gives with a solution of cobaltous sulphate
acidified with acetic acid a yellow precipitate of potassium cobalti-
nitrite, K3Co(N02)6, which is slightly soluble in water. The precipi-
tate may be washed with potassium acetate solution and alcohol.
Potassium cobaltinitrite is decomposed by ammonium sulphide.
The cobaltinitrite is produced only in acidified solutions ; if acetic
acid is not added, a double salt, Co(N02)2,2KN02, is formed.

A reagent for potassium salts is prepared by dissolving 30 gm. of
cobalt nitrate and 50 gm. of sodium nitrite in 150 c.c. of water and
adding 10 c.c. of glacial acetic acid. The salts K2Ag[Co(NO2)6] and
KAg2[Co(NO2)6] are less soluble than K2[Co(NO2)6], hence the addition
of silver nitrate to the above reagent renders it still more sensitive ;
1 part of potassium in 10,000 parts of water may be detected by the
sodium silver cobaltinitrite reagent.

On addition of excess of ammonia to a cobalt salt, a clear solution
is formed which absorbs oxygen from the air, forming complex
compounds known as cobaltammines, which contain ammonia united
with a cobaltic compound, e.g., [Co(NH3)6]Cl3. These show none
of the reactions of cobalt ; the metal is present in the form of complex

1002 INORGANIC CHEMISTRY CHAP.

radicals, e.g., Co(NH3)6. The structure of the cobaltammines will
be considered in connection with Werner's theory of valency (p. 1010).

NICKEL. Ni = 58-21.

Nickel. — The old German miners of Westphalia frequently
obtained a mineral resembling copper ore, from which, however,
no metal could be extracted, and to this the name kupfer-nickel
(i.e., " false-copper," Hiarni, 1694) was applied in derision. In
1750, Cronstadt obtained impure metallic nickel from this ore, the
properties of the element being investigated more thoroughly by
Bergman in 1774.

The chief ores of nickel are the cobalt ore smaltite, (Ni,Co.Fe)As2 ;
white nickel ore, NiAs2 ; kupfer-nickel, or niccolite. NiAs ; nickel
glance, NiAsS ; millerite, NiS ; and the important ores, garnierite,
a double silicate of nickel and magnesium, 2(Ni,Mg)5Si4O]3,3H20,
found in New Caledonia, and pentlandite, (Ni,Cu,Fe)S2, containing
about 2 -5 per cent, of nickel, found at Sudbury, Ontaria. Nickel
ochre, Ni3(As04)2,8H20, also occurs, and the magnetic pyrites of
Pennsylvania contain about 5 per cent, of nickel.

Metallurgy of nickel. — The Sudbury ores, and garnierite, are
roasted, smelted, and bessemerised, yielding monel metal, containing
67 per cent, of Ni, 28 of Cu, and 5 of Mn and Fe, used for sheet metal
work. If monel metal is melted with coke and salt-cake (which form
sodium sulphide) in a basic-hearth furnace, and poled, two strata
separate. The upper layer contains sodium sulphide and cuprous
sulphide, the lower layer is nickel sulphide, NiS. This is roasted to
nickel oxide, NiO, and the latter reduced by heating strongly with
charcoal powder.

Large quantities of nickel are extracted from the Canadian ores
by the Mond carbonyl process, worked at Clydach in South Wales.
In this process (Ludwig Mond, 1895), the roasted ore is leached with
dilute sulphuric acid to remove copper, which is converted into blue
vitriol. The residue is reduced at a temperature below 400° by the
hydrogen contained in water gas. The ferric oxide is not reduced
at this temperature, but nickel oxide forms metallic nickel. The
mass is next passed down a tower provided with shelves, which is
heated at 80°, and carbon monoxide passed through, when volatile
nickel carbonyl, Ni(CO)4, is produced. This is passed through tubes
heated at 180°. Decomposition occurs, and metallic nickel is
deposited : Ni(CO)4 ^ Ni + 4CO, the carbon monoxide passing
back to the volatiliser.

The carbon monoxide is prepared by absorbing carbon dioxide from
flue -gas in a solution of potassium carbonate, heating the bicarbonate
to drive off pure carbon dioxide, and passing the latter over incandescent
coke.

XLIX COBALT AND NICKEL -1003

The total production of nickel in 1909 was about 16,000 tons, 2800 of
which were obtained in England by the Mond Nickel Co.

The metal may be cast. A little magnesium is usually added
before casting, to increase the fluidity and to remove gas-bubbes.

Nickel is refined by electrolytic deposition from a solution of nickel
ammonium sulphate, NiSO4«(NH4)2S04,6H2O, saturated at 20-25°,
a cast nickel block being used as anode and a thin polished sheet of
pure nickel as cathode. The same process is used in nickel-plating,
a thin layer of copper being first deposited on iron or steel goods.
Nickel-plating is fairly easily dissolved by acids, e.g., acetic acid.

Metallic nickel is a metal of a greyish-white colour, of sp. gr.
8 -8, m.-pt. 1484°, very hard and malleable, and capable of taking
a high polish. It is fairly resistant to air but gradually becomes dull ;
it is rendered passive by nitric acid. Nickel is magnetic below 360°.
Finely-divided nickel absorbs 17 times its volume^pf hydrogen. It
decomposes steam at a red heat : H20 + Ni ^ NiO + H2. At
2100°, nickel dissolves carbon, forming a carbide, Ni3C, which
decomposes on coolhig.

Nickel salts. — Nickel dissolves slowly in dilute hydrochloric or
sulphuric acid, evolving hydrogen, but dissolves readily in dilute
nitric acid, a green solution of nickel nitrate being obtained. The
green colour is that of the nickel ion, Ni", and is shown by all the
simple salts of nickel. On evaporation, green monoclinic crystals
of Ni(N03)2,6H2O are deposited. By heating these with concen-
trated sulphuric acid, dissolving the residue in water, and crystallis-
ing, green rhojnbic prisms of nickel sulphate, NiSO4,7H20, separate,
isomorphous* with Epsom salts. If heated with the saturated
solution at 54° these are converted into monoclinic crystals of
NiS04,6H20. Nickel chloride, NiCl2,6H20, is produced by dissolving
the metal in aqua regia, and evaporating. On heating, the green
crystals form the yellow anhydrous salt, NiCl2.

Caustic soda throws down from solutions of nicke.1 salts an apple-
green precipitate of nickel hydroxide, Ni(OH)2, insoluble in excess,
but soluble in ammonia, forming a blue solution, containing two com-
plex cations, Ni(NH3)4" and Ni(NH3)6'', which are derived from
bivalent nickel, and readily lose ammonia .on heating (cf. cobalt).
The ammine salts, e.g., Ni(NH3)6Cl2, Ni(NH3)4SO4,2H2O? may be
obtained in blue crystals. On heating the hydroxide, nickel mon-
oxide, NiO, is obtained as a grey mass, which is also formed on ignit-
ing the nitrate. By gentle ignition of the nitrate a black sesquioxide,
Ni2O3, is formed, which liberates chlorine when dissolved in hydro-
chloric acid. A superoxide, NiO4, is said to be formed by electrolysis,
and a black hydrated dioxide is formed on passing chlorine through
nickel hydroxide suspended in water. A green hydrated peroxide,
NiO2,#H20, or NiO,H202, is precipitated by adding cooled alcoholic

1004 INORGANIC CHEMISTRY CHAP.

potash to a mixture of nickel chloride and H202 cooled to — 50°.
It readily liberates H202 by the action of acids. The black oxide

may be 0=Ni=0, the green oxide Ni<^ | .

Nickel carbonate, NiC03,6H20, is obtained in green crystals by
adding nickel sulphate to a solution of sodium bicarbonate saturated
with carbon dioxide. A green basic salt is precipitated from nickel
salts by sodium carbonate.

Nickel sulphide, NiS, is thrown down as a black precipitate when
ammonium sulphide is added to a nickel salt. It dissolves slightly
in excess of the sulphide, forming a dark brown solution, from which
it is precipitated by boiling, exposure to air, or addition of acids.
Precipitated nickel sulphide readily oxidises in the moist condition
on exposure to air, unless it has been precipitated by boiling a nickel
salt with sodium thiosulphate (cf. p. 828), when it is much denser. It
is insoluble in dilute acids, but dissolves in warm aqua regia. Other
sulphides (Ni2S, Ni3S2, Ni3S4) have been described.

Nickel carbonyl, Ni(CO)4, is a colourless, strongly refracting liquid,
prepared by passing carbon monoxide over reduced nickel at 30°.
It boils at 43-2°, freezes at — 25°, and gives the normal molecular
weight either as vapour or in solution. In the pure state it explodes
at 60°, carbon being deposited : Ni(CO)4 = Ni + 2C02 -f 20. A
mixture of the vapour and air is explosive. Nickel carbonyl is best
prepared under pressure, say. 100 atm., which is favourable to the
right-hand side of the equilibrium: Ni + 4CO ^±Ni(CO)4. At
this pressure, decomposition does not occur even at 250°.

Nickel alloys. — Nickel is used in the manufacture of nickel sted,
of crucibles and tubes, and alloyed with 25 per cent, of copper, for
coinage (U.S.A., Germany, etc.) An alloy of four parts of copper
to one part of nickel is. used for coating rifle-bullets. Nichrom, an
alloy of nickel and chromium, melts at a high temperature, and is
used for electrical resistance heaters. German silver is the alloy
5 copper -}-• 2 nickel -f 2 zinc. Alloys used for resistance coils, etc.,
are :

platinoid : 60Cu + 24Zn + 14Ni + 1-2W.

constantan : 40Ni + 60Cu.

rheostan : 52Cu + ISZn + 25Ni -f 5Fe.

Separation of nickel and cobalt. — These two metals often occur
together in analysis, and their separation may be effected by the
formation of cobalticyanide (p. 1001), nickel forming only the bright
red double salt, Ni(CN)2,2KCN, or K2Ni(CN)4,H2O, easily decom-
posed by acids. This is reduced by sodium amalgam to a lower
cyanide, possibly NiCN. Nitrites form a double salt, Ni(N02)2,
4KNO2, soluble in water, but if calcium salts are present a sparingly

XLIX COBALT AND NICKEL 1005

soluble yellow salt, 2KN02;Ca(N02)2,Ni(NO2)2, similar in appearance
to a cobaltinitrite, may be formed. A solution of nitroso-
/8-naphthol in glacial acetic acid gives a brown precipitate with cobalt
salts, but not with nickel. Characteristic reactions for nickel are
the formation of a scarlet precipitate on addition of a-dimethylgty-
oxime to a solution containing nickel and ammonia or sodium
acetate, and a yellow precipitate on addition of dicyanodiamide and
then caustic potash to an acidified solution of a nickel salt.

Catalytic action of nickel. — Finely-divided nickel, obtained by
reduction of the oxide in hydrogen, acts catalytically in many
reactions involving the absorption of hydrogen. Thus, liquid fats
containing glycerol esters of unsaturated fatty acids such as oleic
(p. 206), B treated with pure hydrogen at 300-400° under pressure
in presence of a little suspended nickel carbonate or borate, absorb
hydrogen and form solid fats, e.g., glyceryl palmitate. This process
is used in the manufacture of margarine.

Estimation of nickel. — Nickel is precipitated from dilute solutions
by adding a slight excess of dimethylglyoxiine, and then ammonia
drop by drop until the liquid smells slightly of ammonia. A bright
red, crystalline precipitate of nickel dimethylglyoxime, Ni(C4H7N2Oo)2,
is formed. This is filtered off, washed, dried, and weighed. The
compound is stable, and sublimes at 120°. The reagent gives a
colour with 0-01 mgm. of nickel.

EXERCISES ON CHAPTER XLIX

1. What are the minerals from which cobalt and nickel are obtained ?
Describe the production of nickel by the Mond process.

2. Describe briefly the properties of the oxides of cobalt and nickel.

3. How are cobalt and nickel separated in analysis ?

4. How is nickel estimated ? In what important respects do cobalt
and nickel differ chemically ?

5. Describe the properties of cobalt and nickel chlorides. What
reactions occur when (a) ammonia is added to a solution of cobalt
chloride, and the liquid is exposed to air, (b) excess of potassium cyanide
is added to a cobalt salt and the liquid is boiled with acetic acid ?

CHAPTER L

THE PLATINUM METALS

Platinum, Pt = 193-6. — The hieroglyphs on an Egyptian box,
discovered at Thebes and dating from 7 B.C., were found by Berthelot
to be composed of an alloy of platinum, iridium, and gold. Specimens
of platinum seem first to have been brought to Europe by Charles
Wood, and it was examined by Bishop Watson in 1750, further
by Margraaf in 1757, and by Bergman in 1777. Platinum foil
and wire were first made in 1772, and in 1806 they were sold
in London, for chemical purposes, at 165. an ounce. The metal now
costs more than £20 per ounce. The normal production of the metal
is about 9 tons per annum. The important Russian deposits in the
Urals were discovered in 1823, and normally supply about 95 per
cent, of the output, the remainder coming from California, Brazil,
Borneo, and Australia, especially New South Wales. It is found in
alluvial sands and gravels, and is separated by washing. The
platinum concentrates consist of metallic grains which, in a specimen
of Russian platinum, had the following composition :

Palla- Osmi-

Platinum Iridium Rhodium dium Gold Copper Iron ridium Sand
76-4 4-3 0-3 1-4 0-4 4-1 11-7 0-5 1-4

Osmiridium is an alloy of osmium and iridium, with small amounts
of other metals :

Osmium Iridium Platinum Rhodium Ruthenium

27-2 52-5 10-1 1-5 5'9

The gold is extracted by amalgamation, and the platinum metals
are digested with aqua regia. Osmiridium remains undissolved. The
solution is evaporated to dryness, and the residue heated to 125°.
Palladium and rhodium form insoluble lower chlorides, PdCl and
RhCl. On treating with water, platinic chloride, PtCl4. and a little
iridium chloride, IrCl4, dissolve. The solution is acidified with
hydrochloric acid and the chloroplatinic acid, H2PtCi6, precipitated
with ammonium chloride as the sparingly ammonium salt,
(HN4)2PtCl6. The iridium remains in solution. On heating
ammonium chloroplatinate it decomposes, leaving spongy platinum.
If this is heated to redness and hammered, the sponge welds into a

1006

CH. L THE PLATINUM METALS 1007

coherent mass of metal. The metal may also be fused in the oxy-
hydrogen flame.

Properties of platinum. — Platinum is a greyish-white metal of
high density, 214, and high melting point, 1753°. It can be
welded at a bright red heat, and may be rolled or drawn into wire.
Very fine wires (Wollaston wires), down to 0-001 mm., are drawn
inside a silver sheath, which can be dissolved off in nitric acid, or by
making the wire the anode in a solution of potassium argentocyanide.
The metal is very resistant, but is attacked by carbon and phos-
phorus at a red heat, becoming brittle.

A smoky flame should not be used with platinum crucibles, nor
magnesium pyrophosphate ignited along with the filter -paper, since in
this case phosphorus is set free. Pure platinum is not attacked on heat-
ing in air, but the modern product loses weight appreciably and becomes
grey and rough after heating. Easily reducible metals such as tin and
lead readily alloy with platinum, causing it to fuse, and compounds of
these metals must not be heated in platinum crucibles with filter -paper.
Caustic alkalies also attack the metal in a fused state, but it is not
attacked by the fused carbonates, nor by hydrofluoric acid. Fused
lithium and magnesium chlorides, and potassium cyanide, attack
platinum.

Pure platinum is not attacked by hot concentrated sulphuric acid,
although the commercial metal dissolves slightly. It is dissolved
by aqua regia on heating, especially if a large excess of concentrated
hydrochloric acid is added. An alloy of platinum and lead dis-
solves in nitric acid, platinum nitrate being formed. On evaporating
the solution in aqua regia, moistening the residue with concen-
trated hydrochloric acid, and re-evaporating, chloroplatmic acid,
H2PtCl6.6H20, is obtained in reddish-brown, deliquescent crystals,
commonly known as " platinic chloride."

Platinum has nearly the same coefficient of expansion as glass and
may be sealed into the latter without causing cracking on cooling. The
wires sealed into electric lamp bulbs were formerly of platinum, but
have been replaced by manganin or Eldred's wire, which has a core of
nickel steel, a jacket of copper, and an outer sheath of platinum. The
metal is used in dentistry and in making jewelry, especially as a setting
for diamonds.

It is used for contacts in electrical apparatus, and in large quantities
as a catalyst in the manufacture of sulphur trioxide and the oxidation of
ammonia. Tantalum has been proposed as a substitute for platinum
in electrical contacts.

Platinum sponge is a grey, porous form obtained by heating ammonium
chloroplatinate. Platinum black is a finely-divided powder obtained by
reducing a solution of chloroplatinic acid by zinc, or with sodium formate

1008 INORGANIC CHEMISTRY CHAP.

solution. These forms are very active catalytically. Alcohol is oxi-
dised by platinum black, on account of its occluded oxygen, to aldehyde,
and a mixture of oxygen and hydrogen is exploded.

Platinised asbestos is made by soaking asbestos fibres, previously
boiled with concentrated hydrochloric acid, in platinic chloride solution,
drying, and heating in a crucible with a little ammonium chloride, or
reducing with sodium formate solution. Colloidal platinum is formed
as a brown solution by causing small electric arcs to pass repeatedly
between platinum wires under water, or by reducing a solution of
platinic chloride with hydrazine in presence of sodium lysalbate, a
protective colloid. The colloidal solution is a catalyst (see H2O2).

EXPT. 341. — Heat a spiral of platinum wire in a Bunsen flame. Turn
off the gas until the wire ceases to glow. Turn on the gas again. The
wire becomes red hot and ignites the gas (see p. 198).

EXPT. 342. — Heat a spiral of platinum wire to redness and suspend it
in a flask containing a little alcohol. The wire continues to glow, and
pungent .vapours of aldehyde, C2H4O, are formed.

Compounds of platinum. — Platinum forms two series of compounds :
the platinous compounds, PtX2, and the more important platinic
compounds, PtX4.

Chloroplatinic acid, H2PtCl6,6H20, the preparation of which has
been described, is a strong dibasic acid ; it gives with silver nitrate
a yellow precipitate of silver chloroplatinate, Ag.2PtCl6 ; the chloro-
platinates of the alkali-metals have been described (p. 797). The
acid therefore gives the ion PtClg" ; on electrolysis this migrates to
the anode, although metallic platinum is deposited on the cathode as
a result of the reducing action of the hydrogen liberated.

On heating potassium chloroplatinate, a residue of platinum
and potassium chloride is left : K2PtClc = 2KC1 + Pt + 2C12.
Ammonium chloroplatinate, (NH4)2PtCl6, leaves a residue of pure
platinum.

Platinic chloride, PtCl4, is obtained as a reddish-brown, crystalline
mass when chloroplatinic acid is heated at 369° in chlorine, or 165° in
hydrochloric acid. At 435°, in chlorine, the greenish -black tri-
chloride, PtCl3, is obtained, and at 580°, brownish-green platinum
dichloride, PtCl2. The dichloride is also obtained by heating the
tetrachloride at 250-300°. Platinum tetrachloride dissolves in
water to form a yellowish -red solution, which appears to contain
a complex acid,' [Pt€l4(OH)2]H2, since it forms a silver salt,
[PtCl4(OH2)]Ag2. Crystals of PtCl4,5H2O may be obtained. Plati-
num dichloride is insoluble in water, but dissolves in hydrochloric
acid to form a dark-brown chloroplatinous acid, H2PtCl4, which is also
obtained by the action of stannous chloride on chloroplatinic acid.

When sodium carbonate is added to chloroplatinic acid solution,
and the residue after evaporation extracted with acetic acid,

L THE PLATINUM METALS 1009

reddish- brown platinic hydroxide, a complex compound, H2[Pt(OH)6],
remains. This dissolves in hydrochloric acid to form
H2[Pt(OHV2Cl4] : silver nitrate gives with the solution Ag2[Pt(OH)6].
On gentle heating, H|.[Pt(OH)6] leaves black platinum dioxide,
Pt02. Platinum trioxide is formed when a solution of potassium
platinate, K2[Pt(OH)6], in caustic potash is electrolysed and the
deposit on the anode, K2O,3Pt03, treated with cold acetic acid ; it
is a brown powder which does not decompose H2O2.

Alkalies precipitate from solutions of platinochlo rides black
platinous hydroxide, Pt(OH)2, probably complex, H2[Pt(OH)4],
soluble in hydrochloric acid. This has no acidic properties ; on
gentle heating it forms black platinous oxide, PtO. Potassium
platinochloride, K2Pt-Cl4, is obtained by warming a paste of potassium
chloroplatinate, K2PtCle, with cuprous chloride. It forms dark red
crystals, used in photography.

Paper is impregnated with a mixture of K2PtCl4 and ferric oxalate.
On exposure to light, the ferric oxalate is reduced to ferrous oxalate,
and if the paper is developed in a solution of potassium oxalate a grey
deposit of platinum is formed on the reduced parts (" platinotype.")

Sulphuretted hydrogen throws down from H2PtCl6 a black precipi-
tate of platinic sulphide, PtS2, soluble in yellow ammonium sulphide
to a dark -brown solution of a thioplatinate, (NH4)4Pt3S6. Platinous
salts give platinous sulphide, PtS.

Potassium iodide does not give with chloroplattnic acid a precipi-
tate of potassium iodoplatinate, but a dark red clear solution. On
heating, this deposits black platinic iodide, PtI4, soluble in alcohol.
When digested with hydriodic acid this forms iodoplatinic acid,
H2PtI6, crystallising in black needles. Platinic iodide decomposes
into iodine and platinum at 130°. Platinous iodide, PtI2, is
obtained as a black powder by heating platinous chloride with
potassium iodide solution.

Complex platinum compounds. — Numerous complex compounds
of platinum are known. The platinammines contain molecules of
ammonia co-ordinated with the metal atom as in the cobaltam-
mines (p. 1001) ; two series exist, corresponding with bivalent and
quadrivalent platinum. Barium platinocyanide, BaPt(CN)4,4H20, is
a lemon-yellow powder used for fluorescent screens in JC-ray work.
Baryta-water and hydrocyanic acid are added to chloroplatinic acid,
the solution warmed, and treated with sulphur dioxide till colourless.
BaSO4 is filtered off and the filtrate crystallised.

Palladium, Pd = 105-9. — When potassium cyanide is added to the
solution of native platinum in aqua regia a pale yellow precipitate of
palladious cyanide, Pd(CN)2, is obtained, which on ignition leaves
metallic palladium (Wollaston, 1803). The metal oxidises superficially
when heated in air, becoming covered with a blue film of monoxide,

3 T

1010 INORGANIC CHEMISTRY CHAP.

PdO. Palladium dissolves in dilute nitric acid, forming palladious
nitrate, Pd(NO3)2, and in aqua regia, forming chloropalladic acid,
H2PdCl6. Potassium iodide throws down from this a black precipitate
of palladious iodide, PdI2, soluble in excess to a brown solution. The
tendency to formation of palladious compounds is noteworthy. The
absorption of hydrogen by palladium has been considered (p. 194).

Osmium, Os = 189'4, and Iridium, Ir = 191-6, — These two metals
are contained in osmiridium (p. 1006). If this is fused with sodium
chloride in chlorine, osmic chloride, OsCl4, volatilises. The solution of
the residue in hydrochloric acid is treated with hydrogen ; platinum
and ruthenium are deposited. When more hydrogen is passed through
the decanted green liquid, iridium is thrown down (Tennant, 1804).

Iridium is very hard, and is used for the tips of gold pens. Iridium
crucibles resist the action of carbon, phosphorus, and aqua regia. The
standard metre of Paris was constructed by Johnson and Matthey, in
London, from an alloy of 90 parts of platinum and 10 parts of iridium.
The same alloy is used, together with pure platinum, in constructing
thermocouples for the measurement of high temperatures. Since
iridium volatilises above 1000°, an alloy of platinum and rhodium is
used at higher temperatures.

When osmium tetrachloride is precipitated with ammonium chloride,
and the ammonium osmicliloride, (NH4)2OsCl6, heated in absence of air,
metallic osmium is left. The metal burns when heated in air or oxygen,
forming the volatile osmium tetroxide, OsO4, commonly called " osmic
acid." This substance has a very irritating odour resembling bromine,
and attacks the eyes. It is easily reduced by organic matter to a black
powder of hydrated dioxide, OsO2 : solutions of osmic acid are used
in microscopy for staining fat globules. The fluoride OsF8 is known.

Ruthenium, Ru = 100-9, and Rhodium, Rh = 102-1, —When the
precipitate of pilatinum and ruthenium obtained as described in the
preceding section is fused with potassium nitrate and caustic potash,
potassium ruthenate, K2RuO4, is formed. The orange-yellow solution
of this, when distilled in a current of chlorine, gives volatile ruthenium
tetroxide, RuO4, similar to OsO4.

Rhodium is contained in the aqua regia solution of the crude platinum
after precipitation with ammonium chloride. If ammonia is added and
the solution evaporated and ignitsd, metallic rhodium is left (Wollaston,
1804).

WERNER'S THEORY OF VALENCY.

Werner's theory of complex compounds. — The formation of so-
called molecular compounds is explained on Werner's^ theory by an
extension of the hypothesis of residual affinity described on p. 252.
In compounds such as H2O and SO3 the principal valencies of the
atoms are saturated, and the molecules are incapable of uniting with

L THE PLATINUM METALS 1011

another univalent atom. By reason of the residual valencies,
however, the two molecules may enter into combination, forming
probably first an association held together by residual valencies :

When association has occurred, the affinities may re-distribute
themselves uniformly, with the formation of radicals, such as S04,
having affinities capable of binding univalent atoms, such as 2H to
form H2SO4. According to Werner, these hydrogen atoms, for
instance, exist outside the sphere of the complex radical S04, and
are therefore easily split off in solution as ions. In this way the
curious behaviour of the chlorine atoms in chloroplatinic acid,
H2PtCl6, is explained. This compound does not give CF ions, but
H' and the complex anion PtCl6". In this case, the six chlorine
atoms are considered to be directly combined with the metal atom in
the central complex, whilst the two hydrogen atoms outside the
nucleus are readily ionisable. In the same way, the ionisable
hydrogen of HC1, when this combines with ammonia, is associated
with the nitrogen atom in the nuclear group NH4 ; the Cl atom,
existing outside the nucleus, is ionisable. The constitution of such
complex compounds as the cobaltammines was formerly explained
(Blomstrand, 1869) by the attachment of ammonia molecules in open
chains to the metal atom, in virtue of the quinquevalent character
of nitrogen : —

e.g., Cof-NH3-NH3-Cl

\NH3-NH3-NH3-C1.

The existence of isomers may be expressed by varying the posi-
tions of the different groups. This theory is no longer accepted, as it
fails to account for many known cases of isomerism.

Alphonse Werner (1893) supposed, on the contrary, that in these
compounds the metal atom is directly united with the non-ionisable
groups such as NH3, N02 (in cobaltinitrites), etc., which form part
of the complex radical, and are not ionised in solution, by what he
calls the supplementary valencies of the metal atom. (These corre-
spond with the " residual valencies " of p. 253.) The principal
valencies of the central atom, which are active in the ordinary
ionisable compounds, e.g., CoCl3, are then free to attach other
ionisable radicals. The complex formed by the supplementary
valencies may be regarded as forming a nucleus outside which the
ionisable radicals are attached. The mode of attachment of the
radicals outside the nucleus is left indefinite by Werner, who places
their symbols outside square brackets enclosing the complex
nucleus, e.g.,

lit m

[Co(NH3)6]Cl3 — [Co(NH3) J- + 3d'.

3 T 2

1012 INORGANIC CHEMISTRY CHAP.

The atoms or radicals in the complex nucleus are said to be co-
ordinated with the metal atom ; the number of such groups is, in
the majority of cases, six, but may be four, as in the complex

platinous compounds, [Pt(NH3)4]Cl2. This number, e.g., 4 or 6,
is called the co-ordination number for the series of compounds.

The valency of the nucleus is equal to the principal positive
valency of the metal atom when the latter is co-ordinated only with
groups, such as NH3 or H2O, which are usually regarded as saturated ;
but if negative radicals such as Cl, which may be considered as
ions, are in the complex nucleus, the positive valency of the metal
atom is reduced by one unit for each such radical present in the
nucleus, and if the number of these radicals exceeds the principal
valency of the metal atom, the complex becomes as a whole negative,
and unites with a corresponding number of positive atoms or
radicals.

E.g., in the compounds of quadrivalent platinum, the complex

[Pt(NH3)5Cl] will have a positive valency of 4 — 1 =3, and will therefore

IV IV

form [Pt(NH3)5Cl]Cl3, whilst the complex [Pt(NH3)Cl5] will have a

valency of 4 — 5 = — 1, and will therefore form [Pt(NH3)Cl5]K. In
the former compound three-quarters of the chlorine, being outside the
nucleus, will be ionisable, and may be precipitated as silver chloride ;
in the latter compound all the chlorine is in the nucleus, is non-ionisable,
and cannot be precipitated as silver chloride :

[Pt(NH3)5Cl]Cl3^±[Pt(NH3)5Cl]'-' + SCI'
[Pt(NH3)Cl5]K — [Pt(NH3)Cl5]' + K\

Isomerism of complex compounds. — Werner's theory predicts
the existence of several kinds of isomers of complex compounds, and
in a great many cases these isomers have been prepared. At first
the theory was in many directions speculative, and encountered
opposition, but with the actual isolation of many of the formerly
hypothetical compounds, the existence of which could not have been
foreseen except by the theory, the value of the latter has become
recognised. At the same time/ it must be admitted that the con-
ceptions used in Werner's theory, e.g., those of " supplementary
valencies," and of the " positions inside and outside the nucleus,"
are vague, but in this respect are at no disadvantage in comparison
with the modern theory of the structure of organic compounds. In
both cases, the cause of valency is unknown.

If the cause of valency is identified with electrical forces, the effect
of substitution of electrically neutral molecules, such as NH3, by ions,
such as Cl, is explained ; the latter exert electrical forces outside the
complex, and the repeated addition of negative atoms to the complex

THE PLATINUM METALS

1013

changes the electrochemical character of the latter in the manner
described.

Seven types of isomerism are possible on Werner's theory : —
(1) Structural isomerism in the nucleus : e.g. :

where en represents ethylenediamine, NH2'CH2'CH2'NH2, with two
supplementary valencies.

(2) lonisation isomerism, in which the positions inside and outside
the nucleus are interchanged, e.g.,

[Co(S04)(NH3)5]Br and [CoBr(NH3)5]S04.

The first splits off a Br' ion ; the second the ion SO4".
The valencies of the two complex radicals are in accordance with
the theory.

(3) Geometrical isomerism. due to the different arrangement of the
atoms and groups in space about the central metal atom : —

(a) In one plane : —

cis-isomer

X R

trans-isomer

A cis-iorm contains adjacent atoms of the same element ; in a
trans-isomeT these are arranged in opposite positions.

(6) Compounds of the type [MeR4X2] can exist in two forms,
which are represented by placing the metal atom (Me) at the centre
of a regular octahedron, with its six supplementary valencies directed
to the six corners. (The possibility that the atoms are arranged in a
plane hexagon is excluded because this would lead to three possible
isomers, whereas only two are known.) The two (univalent)
nuclei of the compounds [Co(NH3)4X2]X are of this type : -

X = Cl, etc.
(negative)

1014

INORGANIC CHEMISTRY

CHAP.

-modifica-

The cis-modifications are distinguished from the
tions by their capacity for ring-formation.

(4) Co-ordination isomerism, depending on the different arrange-
ments of groups in two nuclei in combination :

[Cr(NH3)6].[Cr(SCN)6] and [Cr(NH3)4(SCN)2].[Cr(NH3)2(SCN)4].

(5) Co-ordination polymerism :

[Cr(NH3)3(SCN)3] and [Cr(NH3)5(SCN)]3.[Cr(SCN)6]2.

(6) Hydralion isomerism : the groups NH3, 01, etc., in the nucleus,
may be replaced by water, H20, forming aquo-compounds :

[Cr(NH3)6]Cl3->"[Cr(H20)(NH3)5]Cl3 -> ~~» [Cr(H2O)6]Cl3
ammine compound. aquo-compound.

In such compounds, the ionisable Cl may pass into the nucleus :
[Cr(H20)(NH3)5]Cl3 - [CrCl(NH3)5]Cl2 + H2O

It then ceases to be ionisable. The two green chromic chlorides
(p. 951) are compounds of this type :

[CrCl(OH2)5]Cl2 + H20 and [CrCl2(OH2)4]Cl + 2H20.
The blue modification is [Cr(OH2)6]013.

(7) Optical isomerism : the most convincing argument in favour of
Werner's theory is the existence of optical isomers. These arise when
two compounds have such arrangements of the atoms or groups in
space about the central atom that one structure is the mirror-image
of the other :

•1

eni

iBr

\!NIL

(The bivalent group en engages two valencies of the metal atom, one
axial and one in the plane.)

The existence of optical isomers cannot be detected by the chemical
properties of the compounds : these are identical in both cases. All
the physical properties, such as solubility, density, etc., with one
exception, are also identical. The exception is the behaviour of the
compounds to polarised light. A ray of light is separated by a Nicol's
prism, or other arrangement, into two rays which are complementary in
the sense that the vibrations constituting the light occur in directions
at right-angles in the two cases. Each of these rays, exhibiting a
unilateral vibration, is called polarised. When a polarised ray passes
through a solution of an optical isomer, the plane in which the vibra-
tions occur is rotated, so that if the entering ray is extinguished by a

L THE PLATINTJTM METALS 1015

Nicol prism with its axis at right angles to the plane of polarisation, the
ray after passing through the solution is not totally extinguished,
showing that the plane of polarisation is no longer at right angles to the
axis of the prism. It is found that one isomer rotates the plane of
polarisation to the right, or is dextrogyrous, whilst the other isomer
rotates the plane of polarisation to an equal extent to the left, or is
Isevogyrous. Optical activity is always associated with asymmetric
structure of the molecules of the two compounds, i.e., the spacial con-
figurations of the two molecules are not superposable, but are related
one to the other as an object and its image in a mirror, or as the right and
left hand.

EXERCISES ON CHAPTER L

1. Which elements are included in the group of " Platinum Metals " ?
What are their general chemical properties, and to what uses are they
applied ?

2. Starting with platinum foil how would you prepare : (a) platinum
sponge, (6) colloidal platinum, (c) platinous chloride, (d) chloroplatinic
acid, (e) potassium platinate ? Describe briefly the properties of these
substances.

3. Describe briefly how the metals found in native platinum may be
separated. How would you purify platinum from (a) palladium, (6)
iridium ?

4. What is osmic acid ? How is it prepared from osmiridium, and
for what purpose is it used ? How has the maximum valency of osmium
been established ?

5. Chloroplatinic acid was added to a solution of ammonium chloride ;
the precipitate after ignition left a residue of 0-4752 gm. of platinum.
What weight of ammonium chloride was contained in the solution ?

6. What is Werner's theory of complex compounds ?" Explain the
meaning of : principal valency, supplementary valency, central sphere,
co-ordination number, cis -trans -isomerism, optical isomers, co-ordina-
tion isomerism.

CHAPTER LI

THE RADIO-ELEMENTS AND THE STRUCTURE OF THE ATOM

Cathode rays. — The phenomena of the electric discharge in gases
are described in text-books of physics (e.g., Hadley : "Magnetism
and Electricity for Students." Macmillan). At very low pressure
(0-01 mm.) an electrical discharge proceeds as a blue glow from the
cathode in the exhausted tube, in a course normal to the cathode,
and independent of the position of the anode, producing a green
fluorescence where it strikes the glass. These cathode rays, pro-
ceeding from the cathode, were discovered by Crookes in 1870 ; they
are deflected by a magnet, showing that they are electrically charged.
Perrin was able to demonstrate directly that they were negatively
electrified, and by measuring the deflection produced by mag-
netic and by electric fields, Sir J. J. Thomson (1897) found the ratio
of the charge to the mass of the particles, e/m, to be 1 -2 x 108 cmb.
per gm. ; recent determinations give 1-772 x 108 cmb. per gm. The
corresponding ratio for the hydrogen ion in electrolysis (p. 282) is
F = 9-58 X 104 ; the value for cathode rays is 1850 times this. Of
the two possibilities : (i) the charges are the same, but the mass of
the cathode particle is 1/1850 that of the hydrogen atom ; (ii) the
masses are the same, but the charge on the cathode particle is 1850
that on the hydrogen ion, experiment has decided in favour of the
first. The cathode rays are free negative electrons (p. 281). The

cathode rays have the
Positive same value of e/m, no

matter what is the

material of the electrodes
or the gas in the bulb ;
they are also emitted by
the action of ultra-violet
light on metals, and in
many chemical reactions.

FIG. 424.-Cathode and Positive Rays. electrons being a common

constituent of all atoms.

Positive rays. — If the cathode in the tube is perforated, luminous
rays pass backwards through it (Goldstein, 1886) ; by their de-

1016

CH. LI RADIO-ELEMENTS AND STRUCTURE OF THE ATOM 1017

flections in magnetic and electric fields these are found to consist
of positive particles of atomic size (Fig. 424). The positive rays have
been investi-
gated by Sir
J. J. Thomson,
using the appa-
ratus shown in
Fig. 425. Some
of the particles
were found to
be uncharged.

An electric
discharge is
passed through
the rarefied gas
in a bulb, A,
about 20 cm. FIG. 425. — Thomson's Positive Ray Apparatus,

diameter. The

cathode, C, is an aluminium rod with a rounded end, pierced
by a very fine copper tube, through which a fine pencil of rays
passed between the plates, L and M, which are connected with the
positive and negative poles of a battery, and the pole pieces, P and
Q, of an electromagnet. The cathode is cooled by a water-jacket,
J. The rays are deflected by the combined electric and magnetic
fields, in two directions at right angles to each other, and the
separated pencils of rays then strike a photographic plate. The rays
characterised by definite values of e/m are sorted out into a series of
parabolas on the plate, which are seen in Fig. 426. By measuring
these, the values of m/e and thence, if the charges e are assumed, the
masses, m, of the particles can be calculated. In atmospheric nitrogen
the following results were obtained.

m/e. Nature of Particle

200 Hg+ Mercury atom with single charge.

100 Hg+ + Mercury atom with two charges.

67 Hg+ + + Mercury atom with three charges.

44 CO2 + Molecule of carbon dioxide with single charge.

39 A+ Argon atom with single charge.

N2 + Nitrogen molecule with single charge.

20 Ne + Neon atom with single charge.

15-9 O+ Oxygen atom with single charge.

14 Nitrogen atom with single charge.

12 C+ Carbon atom with single charge.

7 Nitrogen atom with two charges.

X-rays. — When the cathode rays strike the positive electrode, or

1018

INORGANIC CHEMISTRY

CHAP.

anti-cathode, in the bulb, they give rise to a penetrating radiation
which passes outside the tube. This is capable of penetrating
freely through paper, wood, aluminium, and flesh, but is largely
absorbed by lead, platinum, glass, or bone. These so-called X-rays
(Rontgen, 1895) have now been produced sufficiently penetrating to
pass through two inches of steel. They affect a photographic plate,
cause fluorescence when they fall on substances such as barium
platinocyanide, and render a gas conducting or produce ionisation
in the latter, free electrons and positively charged atoms being
formed. For this reason, a gold-leaf electroscope rapidly loses its

charge when exposed to
X-rays, since the sur-
rounding air conducts
away the charge. The
X-rays have been shown
to consist of ether waves
similar to light, but of
much smaller wave-
length. The latter
depends on the composi-
tion of the positive anti-
cathode, or " target,"
bombarded by the
cathode rays and from
which the X-rays proceed.
Bragg's researches on
X-rays and crystals. — For
a long time it was not
possible to obtain diffrac-
tion of X-rays by matter,
since the wave-lengths
are very much smaller
than those of light.
Friedrich, Knipping, and
Laue (1912) showed that
X-rays suffer diffraction

in passing through crystals, and the further work of Sir W. H. Bragg
indicated that they suffered reflexion from crystal surfaces at
definite angles of incidence in the same way as light from a diffraction
grating (Bragg : " X-rays and Crystal Structure," Bell, 1920).

If the primary X-rays are homogeneous, i.e., all of the same
wave-length, the series of directions along which reflexion will
occur are obtained by giving the values 1, 2, 3, ...to n in the
general equation : 2d sin 0 = n X, where X is the wave-length. In
the ordinary diffraction grating, d is the space between the rulings ;
in the case of X-ray reflexion from crystals, Bragg identifies d

FIG. 426. — Positive Ray Parabolas.

LI RADIO-ELEMENTS AND STRUCTURE OF THE ATOM 1019

with the distance between planes in the crystal corresponding with
the densest arrangement of the atoms. These planes correspond
with the symmetry of the crystal (p. 433). It is evident that
an examination of X-ray spectra provides the means of exploring
the atomic architecture of crystals, and in this direction the method
has been applied with great skill and success by W. H. and W. L.
Bragg.

The apparatus used is shown in Fig. 427. The rays from the anti-
cathode of the X-ray bulb are constricted to a narrow pencil by
the lead slits, A and B, and impinge on the crystal, C, mounted on a
rotating arm, V, moving over a graduated circle. The reflected
beams are received in an ionisation chamber, /, also pivoted at the
centre of the X-ray spectrometer, and render the gas contained in the

/_ z

FIG. 428. — Arrangement of
Atoms in Sodium Chloride
Crystal.

Bragg's X-ray Spectrometer.

chamber, usually sulphur dioxide, a conductor of electricity. The
intensity of the current passing through the gas, measured by an electro-
scope, indicates the positions of reflexion from the crystal. The
ionisation occurs with homogeneous X-rays only at certain definite
angles corresponding with the different order of spectra given by the
equation : 2d sin 6 — n\. In the graph of the current against the
angle of incidence, peaks occur corresponding to definite wave-lengths
in the X-rays, and these are repeated as the spectra of different orders
are passed over. In the case of a platinum anti -cathode, for instance
three peaks are found, showing that the X -radiation of platinum is a
mixture of three characteristic wave-lengths. These reappear whatever
the nature of the crystal used for reflexion.

By making use of the principle that the intensity of the radiation
scattered from an atom is proportional to the number of electrons
in the atom, and thus according to the modern theory of atomic

1020

INORGANIC CHEMISTRY

CHAP.

structure (p. 1035) to the atomic weight of the atom, it was possible
to show that the two strong reflexions from potassium chloride were
due to the atoms K and Cl, of approximately equal weight, whilst
the strong and weak reflexions from sodium chloride were due to
the Cl and Na atoms, respectively. In this way the structure of
such crystals was made out to be that shown in Fig. 428, the metal
atoms being represented by dots and the chlorine atoms by circles.

MADAME CURIE

The constituent particles of crystals of salts are not, therefore, the
chemical molecules, such as NaCl, but the atoms (or ions) Na and Cl,
arranged in a definite manner.

Radioactivity. — In 1896, Becquerel found that uranium salts
were capable of affecting a photographic plate through a layer
of black paper, and also of discharging an electroscope.
Thorium compounds were found by Schmidt and by Mine. Curie in
1898 to possess similar properties. The substances were call

LI RADIO-ELEMENTS AND STRUCTURE OF THE ATOM 1021

radioactive, from their property of emitting radiations of the kind
described. In the study of radioactivity the following methods are
available :

(1) The action on a photographic plate.

(2) The phosphorescence produced in platinocyanides, willemite
(zinc silicate), kunzite, and Sidot's spar (zinc sulphide).

(3) The ionisation of gases produced by the rays.

The most convenient is the third method ; the ionisation, which
renders the gas conducting, is detected and measured by the gold-leaf
electroscope (Fig. 429). The strip of gold-leaf, G, is attached to the
vertical rod, R, supported by a horizontal rod, K, insulated on blocks
of sulphur, S, and terminating in a metal plate, B. Below this is a
second metal plate, A, on which the material to be tested is placed.
The motion of the gold-leaf is
observed through a micrometer eye-
piece, the leaf be ng given a charge
through the wire, M, which is
insulated in a sulphur stopper, S,
and can be swung away from the
rod, R, when the latter is charged.
If the substance, C, is radioactive,
the air between the plates A and
B is rendered conducting, owing to
the production of positive and
negative gaseous ions, and the
charge leaks away at a rate which
may be observed by the fall of
the gold-leaf. The electroscope, as
applied to the detection of radio-
active substances, is much the
most sensitive instrument known, since 10~12 gm. of material can
readily be recognised.

Radium. — By means of an electroscope, Mme. Curie found that the
native uranium ore, pitchblende, was more active, for the same
weight of uranium, than a purified uranium salt, and she suspected
that this was owing to the presence of a new element in the ore
which was much more radioactive than uranium. She succeeded
in isolating a trace of an intensely active substance from the pitch-
blende ; this was an impure salt of a new element, radium. It
possessed an activity a million times that of uranium. In highly
purified specimens this activity is found to be doubled.

The separation of the radium from pitchblende is a laborious process.
Radium and barium chlorides are separated by a long series of fractional

FIG. 429. — Gold-Leaf Electroscope.

1022 INORGANIC CHEMISTRY CHAP.

crystallisations (p. 908) ; with the bromides eight crystallisations suffice
for the separation.

An important source of radium compounds is the carnotite of
Colorado (p. 958). This contains 5-10 mgm. of Ra per ton.
The material is boiled with 40 per cent, nitric acid, and the hot
filtered solution deposits barium and radium sulphates on cooling.
These are reduced to sulphides by heating with carbon, the sul-
phides are dissolved in hydrobromic acid, and the salts fractionally
crystallised.

Radium compounds are isomorphous with those of barium ; the
ratio of chlorine to radium in the chloride is 35-2 : 112-15, so that
on the assumption that the formula is RaCl2, the atomic weight of

radium is 224-3.
It is an element
of the group of
alkaline-earth
metals. The
crystals of the
pure salts are
colourless ; if
they contain
barium they are
pink. The solu-
tion in water
evolves oxygen
and hydrogen
continuously,
and the solid
salts ozonise air.
0 X In the dark they

______ C shine with a

• green phosphor-

FiG. 430.— Magnetic Deflection of Rays from Radium. CSCent glow. In

accordance with

the behaviour of the metals of its group, radium sulphate is even
less soluble than barium sulphate, since the element has a higher
atomic weight. In the Bunsen flame radium compounds give a fine
carmine tint, and the spectrum is analogous to those of the other
elements in the group. -

Metallic radium was obtained by Mme. Curie in 1910 by electro-
lysing a solution of the chloride with a mercury cathode, and
separating the mercury from the amalgam by distillation. It is a
white metal, m.-pt. 700°, which rapidly tarnishes in the air, and
decomposes water with evolution of hydrogen.

a-, /?-, and y-Rays. — By interposing sheets of metal foil and
superposing- powerful magnetic fields, in the electroscopic

3ic

Spectrum Tubs

LI RADIO-ELEMENTS AND STRUCTURE OF THE ATOM 1023

method (p. 1021), it was found that radium emits three kinds
of rays (Fig. 430) :—

1. The a-rays : positively charged particles, easily absorbed by
thin metal foil, and having a limited range in air (7-06 cm. when
emitted from RaC).

2. The /3-rays : negatively charged particles, identical with free
negative electrons (p. 281), emitted with speeds approaching the
velocity of light, and often capable of penetrating thin sheets of
aluminium.

3. The y-rays : are not deflected by magnetic fields, and consist-
ing of ether waves identical with very short X-rays (wave-length,
1-3 x 10~7 to 7 x 10~10 mm.), are capable of penetrating several
cm. of lead.

The deflections produced by a magnetic field are seen in Fig. 430 to

be in opposite directions with the a- and /3-rays: the y-rays are un-

deflected. The a-rays have a shorter range than the /3-rays.
The a-rays. — The phosphorescent effects of

radium are mainly due to the a-rays, which,

on account of their relatively large mass,

possess considerable kinetic energy. In the

spinthariscope (p. 267) the impact of each

a-particle on the screen produces a bright

flash and in this way a direct counting of

the particles is possible. The a-rays have

been studied especially by Sir Ernest Ruther-

ford, who found for them the value e/m =

5-07 X 104 cmb./gm., almost exactly half

that for the hydrogen ion in electrolysis.
They may, therefore, consist of atoms of
weight 2 with one positive unit charge,

or atoms of weight 4, i.e., helium atoms,
with two unit charges. By sealing radium
emanation (p. 1025) hi a thin glass tube,

Rutherford and Royds (1909) found that the
a-particles escaped into an outer vacuous

tube fitted with electrodes, .and on passing a discharge through the
latter the helium spectrum was detected (Fig. 431). The a-particle
was thus independently found to consist of an atom of weight 4,
or a helium atom, with two unit positive charges, or as is now believed
a helium atom which has lost two negative electrons. The speed
with which a-particles are emitted by radium is about 2 x 109
cm. per sec., hence the kinetic energy of an a-particle is 1-34 X 10 ~5
erg, or 2-2 x 109 times that of a gas molecule at 0° (p. 268). It is
this large energy which accounts for the phosphorescent effects, and
for the heat evolved by radium, which amounts to 1 18 gm. cal. per gm.

He collects

Eman,

FIG. 431.— Production of
Helium from Jtadium
Emanation.

1024

INORGANIC CHEMISTRY

CHAP.

of radium per hour. The urgent question when the properties of
radium were gradually unfolded was to account for this continuous
production of energy without any appreciable diminution in the
amount of radioactive substance, or of its activity.

The a-particles passing through air produce gaseous ions, which
can act as nuclei for the deposition of moisture. If a particle of
radium is contained in a vessel of air saturated with moisture
and the air is suddenly cooled by expansion, the paths of the rays

FIG. 432. — Tracks of a-rays.

FIG. 433. — Tracks of two
a-rays (enlarged).

become visible in lines of droplets of water, which can be photo-
graphed. In this way C. T. R. Wilson obtained the photographs
shown in Fig. 432. The paths of two single a-rays are shown in
Fig. 433. It will be seen that they end abruptly. The rays must
have passed through several atoms of the gas in their track, without
suffering stoppage or appreciable deflection, but the vertical track
shows a large deflection at its end, and a very small spur is seen
going off in the other direction. The latter probably represents the
track of the atom of gas which has stopped the a-particle ; this has

LI RADIO-ELEMENTS AND STRUCTURE OF THE ATOM 1025

imparted to it a recoil velocit}7, but on account of its larger mass it is
quickly stopped by collision.

The tracks of j8-rays, photographed in the same way, are shown
in Fig. 434. These indicate deflections by collisions with relatively
massive gas particles. The tracks produced by y-rays are shown
in Fig. 435. These are really produced by secondary /3-rays, or
electrons shaken out of the gas atoms by the y-rays.

Radium emana-
tion.— It was soon
noticed that some
kind of gas is con-
tinually evolved
from radium,

which may be
swept away by
a current of air
and is condensed
in a tube cooled
in liquid air. By
measuring the rate
of diffusion of this
gas, called radium
emanation, and by
the direct weigh-
ing of an exceed-
ingly small volume
on the micro-
balance, its atomic
or molecular
weight (on the
assumption that
it is monatomic)
was found to be
220-3. It is an
inert gas belong-
ing to the argon
group. It liquefies
with great sharp-
ness between -- 152° and — 154° ; the liquid boils at — 65°, and
solidifies at — 71°. Under the microscope the liquid is colourless
and transparent, whilst the solid is opaque. The liquid glows with
great brilliancy in a glass tube, with a steel-blue light which at
lower temperatures changes to brilliant orange-red. Ramsay,
therefore, proposed for the gas the name niton, Nt (Latin nitidus =
shining). It has a characteristic spectrum, similar to that of
xenon, and is distinctly soluble in water.

3 u

FIG. 434.— Tracks of /3-rays.

1026

INORGANIC CHEMISTRY

CHAP.

Production of helium from radium emanation. — Ramsay and
Soddy observed that, on standing, the emanation of radium, or niton,
gradually lost its characteristic spectrum, whilst the helium spectrum
appeared. The conversion of niton into helium was definitely
proved by the experiment of Rutherford and Royds already men-
tioned. 3*4 X 1010 atoms of helium are produced from 1 gm. of
radium per second ; the volume of emanation in equilibrium with

1 gm. of radium is 0-585 cu. mm.
This emanation is continually under-
going transformation into helium and
other products, and fresh emanation
is constantly produced from the
radium. Radioactive equilibrium is
therefore not the same thing as
ordinary chemical equilibrium.

The atomic weight of radium is
224 -2 ; the observed density of niton
is 110-6, hence the atomic weight is
221-2. The difference is 3-0, roughly
the atomic weight of helium. The
emanation is therefore produced
together with one a-particle in the
first step in the disintegration of
radium: Ra (224-2) = a-particle
(He = 3-97) + Nt (220-3). Two gases
helium and niton are thus the first
product from the solid radium. The
activity of radium was found to be
quite unaffected by temperature ; it
is the same in liquid air as at a red
heat. In this respect, radioactive
changes differ completely from
ordinary chemical reactions, the
velocity of which is very largely
influenced by temperature.

Theory of atomic disintegration.—
There is no doubt that radium
is an element. It possesses a
definite atomic weight, has a definite

spectrum, and occupies a definite position in the periodic system.
The experiments described above show, however, that radium is
constantly changing into two gases, helium and niton. Each of
these is an element in the same sense as radium. Niton, like radium,
is unstable and produces helium and a solid, which is deposited on
surfaces exposed to the emanation of radium. This solid is called
the active deposit, because it in turn gives rise to other products in

Pm. 435. — Tracks of y-rays.

LI RADIO-ELEMENTS AND STRUCTURE OF THE ATOM 1027

definite stages, each stage in the transformation being accompanied
by the emission of a-rays, i.e., helium atoms, or /2-rays (electrons) and
y-rays. As will be seen later, there are eight changes passed through
in succession from radium to the final product, which is inactive,
and altogether five a-particles and four /3-particles are emitted.
The atomic weight of radium is 224-2, and the five a-particles have
a mass of 5 X 3-97 = 19-9, hence the atomic weight of the final
product will be 224-2 - 19-9 — 204-3. The atomic weight of
lead is 205-55, hence it would seem probable that the final product
of the disintegration of radium is lead. This has been confirmed.

In the above description of the properties of radium, it is assumed
that the atoms of that element and those of some of the products of
change break down and produce new atoms. The puzzle as to the
source of the energy emitted by radium is cleared up by this hypo-
thesis, since it comes largely from the kinetic energy of the swift and
relatively massive a-particles shot from the disintegrating atoms.
The idea of the spontaneous disintegration of atoms was put forward
by Rutherford and Soddy in 1903 ; it follows naturally from the
observed phenomena.

In radioactive changes the transmutation of the elements, so long
but so vainly sought by the alchemists, is proceeding of its own
accord. No human effort can in the minutest detail change any
phase of the process : the rate at which the atoms break down is
unchanged by temperature, by chemical reagents, or by any other
means.

Average life. — An atom of a radioactive element is at any moment
liable to explode, as it were, and give rise to other atoms and
possibly free electrons. The expectation of life of the atoms is
governed by a simple law, discovered by Rutherford. The fraction
of the total number of atoms undergoing disintegration in unit
time is constant ; in other words, the activity diminishes exponen-
tially with the time. The inverse of the fraction disintegrating
per unit time is called the average life, of the element ; it is 1 -443
times the period in which half the atoms have undergone disin-
tegration (half-life). Each radioactive element is characterised by
its average life, which may vary from some million ths of a second,
to millions of years, according as the element is very unstable,
or is more stable, undergoing only slow change.

Radioactivity of uranium. — In 1900 Crookes found that if an
ordinary uranium salt, the radioactivity of which had been dis-
covered by Becquerel in 1896, is treated with ammonium carbonate,
a slight residue is left in which all the photographic activity of the
original salt is concentrated. The solution emits a-rays, which
discharge an electroscope but do not affect a photographic plate,
whilst the residue emits (3- and y-rays, which are photographically
active. The precipitate was called uranium-X ; on standing it

3 u 2

1028 INORGANIC CHEMISTRY CHAP.

became inactive, whilst the solution regained its activity, and
yielded another specimen of uranium-X. Uranium is therefore
capable of growing uranium-X.

Boltwood and Soddy found that radium is produced spontaneously
from uranium, but the change is not a direct one. An intermediate
element, called ionium, was separated by Boltwood from the mineral
carnotite. The same observer also noticed that uranium in
disintegration appears to give out two a-particles, instead of one,
as is the case with most radioactive atoms emitting a-rays. This
suggests that there are two varieties of uranium ; these have
been called uranium-I and uranium-II. Uranium-X also appears to
pass into ionium through an intermediate element known as
uranium -X2. The complete series of transformations of uranium, which
includes that of radium, of which uranium is the parent, is given
in the table below. The periods of average life of the products
are given above the symbols ; the values of the average life are in
years (?/), days (d), hours (h), minutes (m), and seconds (s). The
a-particle, or the helium atom minus two electrons, is denoted by
He ; the /3-ray, or the free electron, by c.

DISINTEGRATION SERIES OF URANIUM.

T65w. ?3xlO%. 10%. 24402/.

He + UXj -> € + UX2 -> € + Un -> He + Io->He+Ra

2440#. 5'55d 4'3w. 38'5m. 28'lw.

Ra -> He + Nt -> He -f RaA -> He + RaB -> e + RaC

l'9m.
28-im (°'°03 Per cent.) He + RaC2 -> e -j- Pb (end-product).

V ? 10-6s. 24y. 7'2rf.

(99-997 per cent.) e + RaCx -> He -f RaD -> e + RaE

196^.

• -> e + RaF -» He + Pb (end-product).
RaC2 and its end-product are members of a branch- chain.

Radioactivity of thorium. — In 1906 Rutherford and Soddy found
that thorium gives off a characteristic emanation, which behaves as
a gas. By adding ammonia to a solution of a thorium salt they
found that the filtrate from the thorium hydroxide contained a
very active substance, to which they gave the name thorium-X.
After a month's time, the thorium-X had completely lost its activity,
whilst the precipitate of thorium hydroxide had recovered exactly
the activity of the original thorium salt, i.e., the activity which had
been lost by the thorium-X. More recent research shows that
Th-X is formed from Th through three intermediate products,
called mesothorium-I, mesothorium-n, and radiothorium. When Th-Cx
is reached, the atoms may disintegrate further in two different
ways. Thirty-five per cent, of the Th-C^ atoms emit an a-particle
formingVTh-D, which then emits a /3-ray, forming lead, whilst 65

LI RADIO-ELEMENTS AND STRUCTURE OF THE ATOM 1029

rent, of the Th-C^ atoms emit /?-rays, forming Th-C2, which then
emits an a-particle, forming lead. Although the net result is the
same in both cases, viz., the emission of an a-particle and a /3-ray,
these changes occur in a different order, and the two final products,
although having the same atomic weight, may have different internal
energies and may be regarded as different atoms. No detectable
rays are emitted by Ms-Thl5 so that the production of Ms-Th2 from
it is called a rayless change.

DISINTEGRATION SERIES OF THORIUM

25X10%. 9-67^. S'Qh. 2'75.v.

Th -> He + MsThj -> MsTh2 -> He + RdTh ->

5'25d. 78s.

He 4- ThX-> He + ThEman.

78s. 0'2s. 15-4A. 87m.

ThEman. -> He + ThA~> He + ThB -> e +ThCr

S7m.

4' 5m.

(35 per cent.) He + ThD -> e + Pb (end-product).

1 (65 per cent.) f -f ThC2 -> He + Pb (end-product).

The actinium series. — Debierne in 1899 separated from the iron
group in the residues of pitchblende from which radium was
prepared another active substance, which he called actinium. The
position of this element in the radioactive series was a puzzle until
Rutherford suggested that it was a branch-chain product, derived
from uranium-II. Ninety-two per cent, of the Un atoms emit
a-particles, forming ionium, but 8 per cent, of them appear to emit
a-particles, forming a different product, known as uranium-Y.
This emits a /?-ray, forming an element recently discovered by Soddy,
called eka-tantalum, which gives rise to actinium by emission of an
a-particle. The final product is a variety of lead.

DISINTEGRATION SERIES OF ACTINIUM.

? 3X lOty

Un ?

11 ^ 22d. ? 103 to 10%. ?30i/.

(8 per cent.) He + UY->e + EkaTa->He + Ac->

28'ld. lQ'4d.

e + RdAc -> He + AcX
16'4rf. 5'6s. O'OOSs. 52'lm.

AcX -» He 4- Ac Eman. -> He + AcA -» He + AcB ->

3'lm. 6.83m.

€ + AcC -> He + AcD -> e + Pb (final product).

It is not yet definitely decided whether UY is derived from Ui or
from Un. It has even been suggested that actinium is derived from a
third isotope (p. 1033) of uranium, which is not a member of the uranium-
radium series but a distinct primary radio-element, called actinouranium.

Atomic numbers. — Kaye (1909) found that a solid element, when

1030

INORGANIC CHEMISTRY

CHAP.

bombarded by a sufficiently rapid stream of cathode rays, emits a
characteristic X -radiation, which may be resolved into a spectrum
by reflexion from a crystal, as explained on p. 1019. Moseley
(1913-4) used a crystal of potassium ferrocyanide and photographed
the spectra of various elements.

The elements (e.g., W, Fe, Cu), or their solid compounds (e.g., KC1),
were used as anticathodes in an X-ray bulb, being mounted on a trolley

FIG. 436.— X-ray Spectra.

inside the bulb so that they could be brought in succession in front of
the cathode by means of a magnet outside.

The spectra consisted in all cases of two main lines, the
frequencies of which decreased in a regular manner as the atomic
weights of the elements increased (Fig. 436). The square-roots of
the frequencies of corresponding lines in the spectra of successive
elements, taken in the order of their positions in the Periodic Table,
when plotted against the number of the element in this table, gave

LI RADIO-ELEMENTS AND STRUCTURE OF THE ATOM 10.)!
TABLE OF ATOMIC NUMBERS AND ATOMIC WEIGHTS.

I 1 Hydrogen

H 1-00

II 2 Helium

He 3-97

III 10 Neon

Ne 20-0

3 Lithium

Li 6-89

11 Sodium

Na 22-82

4 Beryllium

Be 9-0

12 Magnesium

Mg 24-13

5 Boron

B 10-8

13 Aluminium

Al 26-8

6 Carbon

C 11-91

14 Silicon

Si 28-1

7 Nitrogen

N 13-90

15 Phosphorus

P 30-79

8 Oxygen

O 15-87

16 Sulphur

S 31-81

9 Fluorine

F 18-9

17 Chlorine

Cl 35-18

IV 18 Argon

A 39-6

V 36 Krypton

Kr 82-26

19 Potassium

K 38-79

.37 Rubidium

Rb 84-77

20 Calcium

Ca 39-75

38 Strontium

Sr 86-93

21 Scandium

Sc 44-7

39 Yttrium

Yt 88-62

22 Titanium

Ti 47-72

40 Zirconium

Zr 89-9

23 Vanadium

V 50-6

41 Niobium

Nb 92-4

24 Chromium

Cr 51-6

42 Molybdenum

Mo 95-2

25 Manganese

Mn 54-49

43 —

— —

26 Iron

Fe 55-40

44 Ruthenium

Ru 100-9

27 Cobalt

Co 58-50

45 Rhodium

Rh 102-1

28 Nickel

Ni 58-21

46 Palladium

Pd 105-9

29 Copper

Cu 63-07

47 Silver

Ag 107-04

30 Zinc

Zn 64-85

48 Cadmium

Cd 111-51

31 Gallium

Ga 69-5

49 Indium

In 113-9

32 Germanium

Ge 71-9

50 Tin

Sn 117-8

33 Arsenic

As 74-37

51 Antimony

Sb 119-2

34 Selenium

Se 78-6

52 Tellurium

Te 126-5

35 Bromine

Br 79-29

53 Iodine

I 125-91

VI 54 Xenon

Xe 129-2

74 Tungsten

W 182-5

55 Caesium

Cs 131-76

/ O

— —

56 Barium

Ba 136-28

76 Osmium

Os 189-4

57 Lanthanum

La 137-9

77 Iridium

Ir 191-6

58 Cerium

Ce 139-15

78 Platinum

Pt 193-6

59 Praseodymium

Pr 139-8

79 Gold

Au 195-6

60 Neodymium

Nd 143-2

80 Mercury

Hg 199-0

81 Thallium

Tl 202-4

62 Samarium

Sa 149-2

82 Lead

Pb 205-55

63 Europium

Eu 150-8

83 Bismuth

Bi 206-4

64 Gadolinium

Gd 156-1

84 Polonium, or

RaF —

65 Terbium

Tb 157-9

— —

66 Dysprosium

Ds 161-2

67 Holmium

Ho 162-2

VII 86 Niton

Nt 220-6

68 Erbium

Er 166-4

— —

69 Thulium

lTm 167-2

88 Radium

Ra 224-2

70 Ytterbium

Yb 172-1

89 Actinium

Ac —

71 Lutecium

Lu 173-6

90 Thorium

Th 230-31

— . —

91 Uranium X2

U-X2 —

73 Tantalum

Ta 180-1

92 Uranium

U 236-3

1 Thulium has been supposed to be a mixture of two elements, Tmx and Tm2,
but this is not confirmed by the recent work of Urbain on the X-ray spectra.

1032

INORGANIC CHEMISTRY

CHAP.

practically a straight line. If v is the frequency of the line of
longer wave-length ; v0 is a constant (Rydberg's constant) ; N is
the number of the position of the element in the Periodic Table, or
the atomic number, then :

Element . . Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn

Atomic weight .40 — 48 51 52 55 56 59 58-5 63 65

Q . . . 19 21 22 23 24 25 26 27 28 29

N . . . .20 22 23 24 25 26 27 28 29 30

The order of values, of Q is the same as that of the elements in the
Periodic Table, although in some cases (e.g., Co and Ni) the order of
atomic weights is reversed. The atomic numbers of Cl and K,
deduced from the equation above, are 17 and 19, leaving a gap, 18,
for argon, although the latter has an atomic weight higher than
that of potassium. In the table of atomic numbers given on
p. 1031, the values in heavy type have been found experimentally ;
the total possible number of elements from hydrogen to uranium
is 92, five of which still remain to be discovered. There may, of
course, be other elements below hydrogen or above uranium.

The position of radio-elements in the periodic system. — The
position of an element in the periodic system is fixed primarily by
its atomic weight, although in one or two cases (p. 471) the chemical
properties of a pair of elements lead to positions assigned in the
reverse order of the atomic weights. Since the atomic weights of
all the radio-elements have been determined, or (in the majority of
cases) calculated from those of the parent substances by subtraction
of the weight of the helium atoms expelled during the transformation,
it is evident that the positions of these elements in' the periodic
system can be assigned.

If this is done, in conjunction with a consideration of the chemical
properties of the elements, it is found that a general law holds in
all cases. This states (Fajans, and Soddy, 1913) that in an a-ray
change, viz., a radioactive transformation in which an a-particle
is expelled from the atom, the product generated falls into a group
of the periodic system two places lower than that to which the parent
substance belongs. In a /3-ray change, on the other hand, viz., one
in which an electron is expelled from the atom, the product falls
into a group one place higher than that of the parent substance.

Thus, the expulsion of an a-particle from the atom of radium, an
element of the second group, leads to the formation of niton, an inactive
gas of the zero group ; the expulsion of a /3-particle from Ra D, an
element of Group IV, leads to the formation of Ra E, an element of
Group V.

The whole series of radio-elements and their transformations are

FIG. 437. — Periodic Arrangement of Elements.

LI RADIO-ELEMENTS AND STRUCTURE OF THE ATOM 1033

shown in Fig. 437, in which their positions in the periodic table are
evident. One of the most striking results appearing from this
figure is that more than one element occupies the same place in
the system (p. 462). The different elements occupying the same
position in the table are called isotopes ; they are inseparable one
from another by chemical processes, and behave chemically as
identical elements. They can be distinguished, however, by their
radioactive properties, in particular by the rate of disintegration,
and by the nature of the elements from which they are derived, or
of the products to which they give rise. They are also differentiate
by their atomic weights and by their densities, since their atomic
volumes are identical. Soddy (1918) has pointed out that two types
of isotopes exist :

1. Those of different atomic weights, which are products of different
disintegration series, such as the varieties of lead (p. 462). These are
known as heterobaric isotopes.

2. Those of the same atomic weight, produced by the successive
expulsion of an a-ray and a /3-ray in different orders, i.e., in one case an
a-ray is lost first, and then a /3-ray, whilst in the second case a /3-ray is
first expelled, and then an a-ray. These are known as isobaric isotopes.
Ra D and Pb from Ra C2 are isobaric isotopes.

Elements occurring in different positions in the periodic system
are called heterotopes ; they are separable by chemical means. They
may have different atomic weights, when they are called heterobaric
heterotopes, or the same atomic weight, when they are called isobaric
heterotopes.

It has been found that the molecular solubilities of compounds of
isotopes are identical. Thus, the solubilities of common lead nitrate
and of uranio-lead nitrate are 1-7993 and 1-7991 gin. mol. per litre,
respectively. The actual weights of lead per 100 gm. of water are
37-281 and 37-130, substantially in the ratio of the atomic weights.

The spectra of isotopes have so far been found to be identical, both as
regards the ordinary light spectra and the high frequency, or X-ray
spectra. The X-ray spectra of ordinary lead and of uranio-lead were
found to be identical within the error of 0-0001 Angstrom unit. The
infra-red spectra have been neglected, and in one case it has been thought
that a minute difference has been detected in the X-ray spectra. It is
also possible that isotopes may have different vapour pressures.

Fig. 437 shows that the radioactive series extends over twelve
places in the periodic table, the places occupied by the halogens
and the alkali-metals, viz., Groups I and VII, being entirely skipped.
It is perhaps worthy of note that just these two groups contain the
strongest electropositive and the strongest electronegative elements
known. It is also noteworthy that, in order to preserve the relation
described by Fajans, viz., the passage into the next group but one
lower by the loss of an a-particle, and into the next higher group

1034 INORGANIC CHEMISTRY CHAP.

by the loss of a ^-particle, the group of transitional elements,
Group VIII, has to be omitted altogether. This indicates that some
place should be found for these elements in the rest of the table, but
so far none of the attempts to do this have been successful.

In the ten occupied places in the last two periods there are
forty-three distinct types of atoms, characterised by specific
radioactive properties, but these represent only ten chemically
different substances. The chemical and spectroscopic characters of
seven of these, viz., Tl, Pb, Bi, Nt, Ra, Th, and U, have been firmly
established, and the places occupied by them accommodate all but
nine of the known radio-elements.

The tendency of workers on radioactivity is to regard isotopes as
different elements ; since, however, they are identical in chemical
properties, it has been suggested by Paneth (1916) that they should
be regarded as varieties of elements, the latter being substances
which cannot be simplified by chemical means. Fajans, on the
other hand, adopts the view that they are different elements, and
would define an element as a substance which cannot be separated
by any chemical or physical means into simpler constituents, and
cannot be recognised as a mixture of other substances (e.g., a
mixture of isotopes).

It is evident how deeply these discoveries reach into the fundamental
conceptions of chemistry. The position has been eloquently put by
Soddy, to whom so much of this fascinating work is due : " Nemesis,
swift and complete, has indeed overtaken the most conservative con-
ception in the most conservative of sciences. The first phase robbed
the chemical element of its time-honoured title to be considered the
ultimate unchanging constituent of matter ; but since its changes were
spontaneous and beyond the power of science to imitate or influence
to the slightest degree, the original conception of Boyle, the practical
definition of the element as the limit to which the ultimate analysis of
matter had been pushed, was left almost unchanged." After pointing
out that, during the last century, the atom and the element were
regarded as synonymous, related as the singular to the plural, and that
the atoms of any one element were considered to be identical in every
respect, Soddy proceeds to say : " The second phase in the develop-
ment of radioactive change has now negatived each and every one of
the conceptions of last century that associated the chemical element
with the atom. The atoms of the same chemical element are only
chemically alike. Different chemical elements may have the same
atomic mass, the same chemical element may have different atomic
masses, and, most upsetting of all, the atoms of the same element may
be of the same mass and yet be an unresolvable mixture of funda-
mentally distinct things."

The age of the earth. — Calculations by Lord Kelvin on the assump-

LI RADIO-ELEMENTS AND STRUCTURE OF THE ATOM 1035

tion that the earth consisted originally of a sphere of incandescent
matter which has cooled by radiation into space, showed that the
time required to arrive at the present condition was much shorter
than the period indicated by the geological deposits and fossils.
The presence of radioactive material in the earth has modified this
calculation, since the heat evolved in its disintegration would make
the cooling process much slower than would otherwise be the case.
The source of energy in the sun may also be due partly to radioactive
changes ; although radium has not been detected in the solar
spectrum, the presence of helium, one of the products of disintegra-
tion of radium, suggests the possibility of radioactive changes.

The structure of the atom. — The sudden and often large deflection
of the a-particle at the end of its track, shown in Fig. 433, indicates
that its positive charge must have approached very close to some
positive charge in an atom of gas, in such a way that a large repulsive
force arises between the two like charges. As the a-particle must
have passed through several atoms without deflection before it is
finally arrested, this positive atomic charge must be concentrated
in a volume of small dimensions compared with the volume of the
atom. Calculation shows that the two charges must have approached
within a distance of 10~13 cm., i.e., less than the radius of the electron
and many times smaller than the radius of an atom, which is of the
order of 10~8cm. The atom is electrically neutral, so that in
addition to the positive nucleus there must be electrons. The
simplest assumption as to the structure of the atom is that it consists
of a very small positive nucleus surrounded by one or more electrons
revolving in circular orbits about the nucleus, the diameter of the
electronic orbit being of the order of the diameter of the atom.
The rest is empty space. By far the greater part of the atom is
therefore space.

The electrons have a very small mass, so that the mass of the atom
must be concentrated in the positive nucleus. The mass associated
with an electric charge is inversely proportional to the radius of the
charged body ; the very small radius of the positive nucleus implies
therefore a large mass. It is not necessary to assume that any part
of the mass of the atom is not that associated with the positive charge
on the nucleus and the small mass of the orbital electrons. The
whole mass is then regarded as electric, and matter is looked upon
as an aggregate of electric charges.

The above theory of the structure of the atom is due to Rutherford
(1911). Since no positive charge has been isolated less than the
mass of the hydrogen atom, the latter is assumed to consist of one
unit positive nuclear charge, with one electron revolving around it
to make the whole electrically neutral (Fig. 438). The helium
atom would then contain a nucleus carrying four unit positive
charges and two electrons. The nucleus in this case is identical

1036 INORGANIC CHEMISTRY CHAP.

with the a-particle ; the positive nucleus of the hydrogen atom, or
the unit positive charge, is the hydrogen ion. The series of atomic
numbers suggests that successive atoms, counting from helium,

_.< _ have nuclei containing one additional

,.-*" ""*%„ charge for each step in atomic number,

i^ ® — — C) All atoms contain the unit positive

v%-^ ,.*' charge, i.e., the hydrogen ion, as the

* >---** basis of the nucleus . The helium nucleus

FIG 438.— structure of appears to be a secondary nucleus of great

.Hydrogen Atom. * \ .,.,

stability.

In chemical changes only the orbital electrons are disturbed.
The ionisation of potassium, for instance, implies a loss of one
orbital electron, the nucleus remaining unchanged. The vibrations
of orbital electrons, or their shift from one orbit to another, is sup-
posed to give rise to the radiation emitted by the atom, i.e., to its
spectrum. It is only in radioactive changes, when a- and /2-particles
are emitted, that disruption of the nucleus occurs. The electrons
of /8-rays are assumed to come from the nucleus, so that the latter
may contain negative electrons as well as positive charges. There
must always be a net positive charge on the nucleus to maintain
neutrality with the orbital electrons. The a-particle consists of
four hydrogen nuclei plus two nuclear electrons, since it has a net
positive charge of two units. The helium atom has two orbital
electrons in addition.

The structure of molecules. — The views on the structure of molecules,
apart from those in crystals studied by the Braggs, are less definite than
those relating to atoms. Debije considers
that in the formation of the hydrogen
molecule the two orbital electrons of the
atoms are displaced. The two circular
currents represented by the rotating

electrons attract, whilst the positive nuclei ®

repel, each other. The orbits of the

electrons approach more rapidly than the

nuclei and finally coalesce. When this

occurs the hydrogen molecule is produced

(Fig. 439). This consists of a pair of

electrons rotating in a plane at right angles Fig 439.— Structure of

to the line joining the two nuclei.

EXERCISES ON CHAPTER II

1. How are positive rays produced ? Of what do they consist ?

2. Explain how X-rays have been used to study the internal structure
of crystals.

Li RADIO-ELEMENTS AND STRUCTURE OF THE ATOM 1037

3. What is the atomic number of an element ? How are these
determined ?

4. Give a brief account of the phenomena of radioactivity. How was
radium discovered and how is it separated from its ores ?

5. What kinds of radiations are emitted from radium ? How are
they distinguished ?

6. Give a brief account of the disintegration theory of radioactivity.
On what experimental evidence is it based ?

7. What are isotopes ? How is their formation explained ?

8. Describe briefly the position of the radio -elements in the Periodic
System.

9. What are the modern views on the structure of the atom? On
what experimental evidence are they based ? Draw diagrams indicating
the structure of (a) the hydrogen atom, (b) the hydrogen molecule,
(c) the helium atom, (d) the a-particle, (e) the lithium atom.

PERIODIC SYSTEM 01

SERIES

GROUP 0.

GROUP 1.

GROUP JI

GROUP III.

GROUP IV.

—

R20
RH

1£

R208

R204
RH4

He 3-97

H I'OOO

Be 9-00

B 10-8

C 11-910

Li 6-89

Ne 20-0

Na 22-82

Mg24-13

Al 26-8

Si 28-1

A 39-6

K 38-79

Ca 39-75

Sc 44-7

Ti 47-72

Cu 63-07

Zn 64-85

Ga 69-5

Ge 7! -9

Kr 82-26

Rb 84-77

Sr 86-93

Yt 88-62

Zr 89-9

^ Ag 107-04

Cd 111-51

In 113-9

Sn 117-§!

Xe 129-2

Cs 131-76

Ba 136-28

La 137 -9 (and
12 other
elements of
RareEarths)

Ce 139-15

Au 195-6

Hg 199-0

Tl 202-4

Pb 205-55

Ra-C2
Ac-D
Th-D

Pb ex Ra-C«
Pb ex Ra-F
Pb ex Ac-D
Pb ex Th-Ci
Pb ex Th-D
Ra-B
Ra-D
Ac-B
Th-B

Nt or Ra-
Eman.
220-6

Ac-Eman
Th-Eman.

—

Ra 224-2

Ac-X

Ms Th,
Th-X

Ac?
MsTh2

Th/230-31

U-X,
lo
U-Y
Rd-Ac
Rd-Th

1038

THE ELEMENTS.

GROUP V.

GROUP VI.

GROUP VII

R205
RH3

R206
RH2

R207
RH

N 13-897

O 15-87

F 18-9

P 30-79

S 31-81

Cl 35-18

GROUP VIII.

V 50-6
As 74-37

Cr 51-6
Se 78-6

Mn 54-49
Br 79-29

Fe 55-40 Co 58-50 Ni 58-21

Nb 92-4
Sb 119-2

Mo 95-2
Te 126-5

I 125-91

Ru 100-9 Rh 102-1 Pd 105-9

Ta 180-1

Bi 206-4

Ra-Cj
Ra-E
Ac-C
Th-C

W 182-5

•

Po or Ra-F
Ra-A
Ra-C
Ac-A
Th-A
Th-Ci

—

Os 189-4 Ir 191-6 Pt 193-6

•

EkaTa
U-X?

U-i 236-3
U-n

1039

ANSWERS TO EXAMPLES

Chapter I
4. 625 : 1.

Chapter IV
4. 2-185gm.

Chapter V

1. 45-79 c.c.; 0-0654 gm. 2. 350-8 cu. ft. 3. 383-08 mm.

6. 53-42. 7. 22-42. 8. (i) l-204kgm. ; (ii) 83-3 per cent.

9. 0-0943 gm.

Chapter VI
4. 0-086 gm. per litre. 9. 339-6 c.c.

Chapter VII

4. 32-64 (density of H = 0-08987 gm. per litre).

5. Ag. = 107-92, S = 16-032.

Chapter VIII
4. 139-48 gm. 5. 157-88 litres.

6. 228 gm. ferrocyanide, 362 gm. sulphuric acid, and 19-7 gm. water.

Chapter IX

3. 578-44 gm.

4. 1-798 ; 2-326 gm. per litre ; C2N2 ; 2 litres of CO, 1 litre of N2.
6. C12 21-86 lit. ; CO2 22-09 lit. ; NH3 21-92 lit.

8. At. wts. Cl = 221-6 ; H = 6-3 ; Mol. vols. = 140 lit.

10. At. wt. 236-4. XC14. 13. A = 82-6 ; 7 = 0-5375.

14. 27-818.

Chapter X

9. (a) 5-484 litres, (b) 8-954 litres.

10. 24-45 c.c. at S.T.P. ; O2 = 47-24 per cent., N2 = 52-76 per cent, by vol.

Chapter XI

5. 3-79 c.c.

Chapter XII
3. 0-026 Ib.

1040

ANSWERS TO EXAMPLES 1041

Chapter XIII
2. 30-2 gm. C12 ; 5-03 litres.

Chapter XV

8. 1 : 0-586.

Chapter XVI

2. 458 ampere hours. 7. 0-0853 gm.

10. 93-8 c.c. ; 0-0902 gm.

Chapter XVII

5. 260-7. Probably undissociated. 6. 364.

7. 88-3 per cent. ; 87-4. 10. 1757 ; (H2WO4)7.

Chapter XVIII

6. 7-5 c.c.

Chapter XIX
5. 29-53 grn.

Chapter XXI

9. Nil ; 15 c.c. of O2.

Chapter XXIII
1. Se = 78-63. 2. Al = 26-89.

5. Sp. heat of S = 0-163. 6. 0-53.

11. Isomorphous mixture 1-44 FeCO3 + MnCO3 ; or (Fe, Mn) CO3.

Chapter XXVIII

6. 10-5 c.c. NH3, 9-5 c.c. N2.

Chapter XXXII

8. Mol. wt, =77-5, gas is AsH3.

Chapter XXXIII

15. 12-5 H2, 7-5 CH4, 80 N2 by vol.

16. CH4 35-8 c.c., C2H6 10-4 c.c., H2 7-3 c.c.

Chapter XXXV

7. 10-94.

Chapter XLU

4. 1-0186 x 96,000 X 2 = 195,580 joules = 46,710 gm. cal.

6. 12,343 gm. cal. (Heat of formation in solution 13,200 gm. cal.)

Chapter L

5. 0-2605 gm.

3 x

INDEX

Absorptiometer, 95
Absorption coefficient, 96
Acetaldehyde, 680
Acetylene, 677

Acids, 134, 771 ; chlorides, 515; con-
ductivity of, 288; dibasic, 229;
monobasic, 229; oxy-, 517;
preparation of, 385 ; properties
of, 771 ; strengths of, 184, 288,
735, 771
Acid, acetic, 680

allotelluric, 533

amidosulphonic, 597

antimonic, 933, 935

antimonious, 934

arsenic, 646, 654

arsenious. 653

azulmic, 716

benzene sulphonic, 511

boracic (see boric)

boric, 733 ; tests for, 738

bromic, 401

bromous, 401

carbamic, 708

carbonic, 689

Caro's, 519

chlorantimonic, 937

chlorantimonious, 936

chlorauric, 835

chloric, 383

chlorobismuthous, 944

chlorochromic, 956

chlorosulphonic, 515

chlorous, 381

chromic, 947, 953

citric, 771

cobaltic, 999

cobalticyanic, 1001

cyanic, 718

dichromic, 953

dinitropyrosulphuric, 596

disilicic, 746

PARTINGTON'S INORQ. CHEM.

Acid, dithioiiic, 523
ethionic, 677

ethylsulphuric, 496, 675, 677
ferric, 993

ferricyanic, 995 '

ferrocyanic, 994
fluoboric, 737
fluosilicic, 751
formic, 674, 709
glycollic, 710
glyoxylic, 710
graphitic, 664
hexathionic, 524
hydrazoic, 541, 558
hydriodic, 402, 408
hydriodostannous, 916
hydrobromic, 397
hydrochloric, 148, 218, 221, 229,

238

hydrochloromercuric, 874
hydrochloroplumbic, 925
hydrochlorostannic, 917
hydrocyanic, 716
hydrofluoaluminic, 898
hydrofluoric, 419
hydrofluosilicic, 751
hydrographitic, 665
hydroxylamine disulphonic, 506,

597

hypo bromous, 401
hypochlorous, 369, 374
hypoiodous, 412
hyponitrous, 554, 593
hypophosphoric, 641
hypophosphorous, 618, 642
hyposulphurous, 525
iodic, 405, 412
iodobismuthous, 944
lactic, 771
malic, 771
malonic, 710
manganic, 967

1043 3x2

1044

INDEX

Acid, rnellitic, 664

molybdic, 957

nitric, 565 ; ac*bion of, on metals,
570 ; manufacture of, 572

nitrilosulphonic, 597

nitrohydroxylamic, 594

nitrosodisulphonic, 506

nitrosoferricyanic, 996

nitrososulphuric, 505, 595

nitrosulphonic, 590

nitrous, 542, 584 ; constitution of,
537

nitrylsulphonic, 506

osmic, 1010

oxalic, 710

oxamic, 716

pentathionic, 524

perboric, 738

percarbonic, 693

perchloric, 383

periodic, 413

permanganic, 967

pernitric, 593

perphosphoric, 634

persulphuric, 518

phosphatic, 641

phosphimic, 625

phosphomolybdic, 597

phosphoric, 634 ; constitution, 635 ;
titration, 63,1 ; meta-, 628, 632 ;
ortho-, 629; glacial, 629; pyro-,
632

phosphorous, 619, 635

phosphotungstic, 958

picric, 569

plumbic, 924

prussic, 717

pyroligneous, 666

pyrosulphuric, 501

selenic, 530

selenious, 528, 530

silicic, 744

silicofluoric, 751 t

silicon formic, 750

silicon meso -oxalic, 750

silicon oxalic, 750

stannic, 914, 918

sudoric, 790

sulphonic, 501

sulphonitronic, 506

sulphovinic, 677

sulphoxylic, 526

sulphuric, 499 ; concentration of.
506 ; estimation of, 512 ; fum-
ing, 501 ; manufacture of, 499 ;
properties of, 509 ; purification
of, 508

Acid, sulphurous, 493, 496
tartaric, 771
telluric, 562
tellurous, 532
tetrathionic, 522
thioantimonic, 938
thioarsenic, 656
thioarsenious, 656
thiocarbonic, 714
thiolcarbonic, 515, 709
thioncarbonic, 715
thiophosphoric, 637
thiostannic, 919
thiosulphuric, 520
trithionic, 523
tungstic, 657

Actinium, 1029 ; series, 1029
Actinometer, 235
Actinouranium, 1029
Active deposit, 1026 ; mass,

350

Adsorption, 667
Acs cyprium, 805
Affinity, 344, 805, 879 ; series, 886
Agates, 744
Air, composition of, 535, 537 ; fixed,

773; liquid, 172
Alabandite, 961
Alabaster, 846
Albite, 746
Albumin, molecular weight of,

316

Alchemists, 832
Alchemy, 28, 31, 764, 832,

1027
Alcohol, 677, 680; absolute,

845

Algin, 405
Algulose, 405
Aliphatic compounds, 659
Alizarin red, 363

Alkali, marine-, 773 ; -metals, 770 ;
vegetable-, 773 ; volatile-, 773 ;
-waste, 778, 847

Alkalies, 772 ; manufacture of, by
electrolysis, 296 ; by Leblanc
process, 777 ; by Solvay process,
782

Alkaline earth metals, 838
Allotropy, 114
Alloys, 764 ; binary, 766 ; freezing

point curves of, 766
Alsatian potash deposits, 791
Alum, ammonia, 900 ; burnt, 900 ;
indium and gallium, 900 ; neutral
899 ; potash, 900, ; Roman,
899; -shale, 899

INDEX

1045

Alumina, 776, 891, 894

Aluminium, 890 ; acetate,

arsenide, 648 ; bromide, 898
bronzes, 893 ; chloride, 897
fluoride, 898 ; hydroxide, 892,
895 ; iodide, 898 ; manufacture
of, 892 ; nitrate, 900 ; nitride,
544, 900 ; oxide, 890 ; peroxides,
896 ; properties of, 893 ; rectifier,
894 ; resinate, 847 ; sulphate,
898 ; sulphide, 900

Alundum, 895

Alunite, 899

Amalgamation process, for silver,
821

Amalgams, 765

Amatol, 801

Amethyst, 741 ; oriental, 894

Amicrons, 8

Amino- group, 544, 547

Ammonia, 542, 773 ; by-product,
550 ; composition of, 548 ;
-liquor, 550 ; oxidation of, 575 ;
preparation of, 545 ; properties
of, 546 ; -soda process, 781 ;
-stills, 551 ; synthetic, 190, 543

Ammonium, 770; amalgam, 798;
bicarbonate, 802 ; bromide, 800 ;
carbamate, 802 ; carbonate, 801 ;
chloride, 542, 799 ; chloroplum-
bate, 925 ; chromate, 955 ;
cyanate, 718; dichromate, 539,
955 fluoride, 800 ; hydroxide,
547 iodide, 800 ; molybdate,
957 nitrate, 801 ; nitrite, 539,
586 oxide, 547 ; peroxide,
547 phosphomolybdate, 631 ;
sesqui carbonate, 802 ; sulphate,

542,

552, 800; sulphides, 800;

sulp lite,' 801 ; thio carbonate,
714 thiostannate, 916

Ampere, 279, 282, 825

Amphoteric electrolyte, 776

Analysis, 26 ; spectrum, 755

Anatase, 929

Andalusite, 746

Anglesite, 920

Angstrom unit, 756

Angus Smith's compound, 985

Anhydrides, acid, 135

Anhydrite, 846

Anions, 278

Anode, 278

Anthracite, 670

Antichlor, 369, 522

Antifriction metal, 934

Antimonial lead, 934

Antimoniates, 935

Antimoniuretted hydrogen, 938

Antimony, 607, 764, 932 ; alloys,
934 ; allotropic forms, 933 ;
chlorides, 933 ; estimation, 939 ;
halogen compounds, 936 t
hydride, 938 ; nitrate, 934 ;
oxides, 932, 934, 935 ; sulphate,
934 ; sulphides, 932, 937 ; ver-
milion, 937

Antimonyl, 940

Apatite, 608

Aqua regia, 594

Aqua tofani, 651

Aqua vieja, 405

Aragonite, 840

Argentite, 819

Argentum cornu, 827

Argentum vivum, 868

Argon, 600

Argyrodite, 929

Armour plate, 983

Aromatic compounds, 659

Arsenates, 654

Arsenic, 607, 644 ; dioxide, 653 ;
halogen compounds, 650 ; hy-
dride, 647 ; hydroxy chloride,
651 ; iodides, 651 ; pentachlor-
ide, 651 ; pentafluoride, 650 ;
pentoxide, 654 ; solid, 650 ;
sulphides, 12, 655 ; trichloride,
650 ; tribromide, 651 ; tri-
fluoride, 650 ; thio-acids of, 65$ ;
tribxide, 644, 651 ; sulphates,
647, 652 ; tests for, 648, 652 ;
white, 645

Arsenical iron, 644 ; -pyrites, 644 ;
-nickel, 644

Arsenite, 644

Arsenites, 653

Arseniuretted hydrogen, 647

Arsine, 647

Asbestos, 747, 855

Association, 150, 317

Atacamite, 814

Atmolysis, 170

Atmosphere, 535

Atomic disintegration, 1026

Atomic heats, 146, 425, 455 ; at low
temperatures, 427 ; numbers,
452, 1029, 1031 ; theory, 127 ;
volumes, 453, 456

Atomic weights, definition of, 128 ;
determination of, 143, 425 ; from
isomorphism, 445 ; from Periodic
Law, 468 ; from specific heats,
430 ; table of, 145

1046

INDEX

Atoms, absolute weight of, 128,
268 ; mass of, 1035 ; mode of
linkage, 391 ; spontaneous dis-
integration of, 1027 ; structure
of, 1016, 1035

Augite, 855, 890

Auricome, 338

Aurodiamine, 836

Aurous and Auric compounds,
see Gold

Autocatalysis, 383

Autoxidation, 166, 342, 833

Available chlorine, 379 ; oxygen,
970

Avogadro's constant, 262, 266,
268, 313 ; hypothesis, 138 ;
law for solutions, 311

Axis, brachy-, 438 ; clino-, 440 ;
lateral-, 438 ; macro-, 438 ;
ortho-, 440 ; vertical, 438

Azoimide, see Hydrazoic acid

Azurite, 805, 814

Bacteroids, 576

Baking powder, 688

Barff process, 985

Barilla, 777

Barium, 852; bromate, 402; car-
bonate, 851 ; chlorate, 383 ;
chloride, 851 ; chromate, 955 ;
cyanamide, 716; cyanide, 716;
dithionate, 523; ferrate, 993;
hydroxide, 851; iodate, 413;
me tabor ate, 735 ; nitrate, 851 ;
nitrite, 585 ; oxide, 851 ; percar-
bonate, 333; periodats, 413;
permanganate, 967 ; peroxide,
333, 852 ; salts, 851 ; sulphate,
853 ; sulphide, 851, 877 ; sulphite,
495

Barysilite, 746

Baryta, 850, 852 ; -water, 852

Barytes, 850

Bases, 134. 771, 776

Bauxite, 891, 895

Becher, 38

Beckmann thermometer, 301

Bedil, 912

Bell metal, 810

Benzene, 680

Bergman, on affinity, 345

Berthelot, partition law, 98 ; gas
equation, 269

Berthollet, on affinity, 345 ; and
Proust, 111 ; on chlorine, 221,
368

Beryl, 746, 854

Beryllium, 469, 854 ; compounds,

854
Berzelius, volume theory, 139 ;

dualistic theory, 274
Bessemer process, 979
Betts process, 921
Bischof process, 921
Bismuth, 607, 764, 940 ; bromide,

carbonate, 943 ; chloride,

chromate, 955 ; colloidal,

dioxide, 942 ; fluoride,

hydride, 944 ; hydroxide,

iodide, 944 ; magpie test

942 ; metaphosphate, 943 ;

944
944
941
944
941
for,

nitrate, 941 ; oxides' 942 ; ores,
940 oxychloride, 944 ; phos-
phate, 943 ; sulphate, 943 ;
sulphide, 9*3

Bismuthyl radical, 942

Biscuit, 901

Bittern, 220

Black, J., research on alkalies, 773

Black ash, 778

Blackband ironstone, 974

Blacklead, 663

Blagden, 54, 103

Blast furnace, 807, 975

Bleaching, by chlorine, 223, 269 ;
by hydrogen peroxide, 338 ; by
sulphur dioxide, 494

Bleaching powder, 376

Blomstrandite, 907

Blood, 697, 970

Blue-fire, 938 ; -stone, 812 ; vitriol,
812

Bodies, classification of, 26

Bog iron ore, 974

Boiling point, molecular elevation
of, 304

Bolometer, 759

Bone, 609 ; -ash, 609

Boracite, 732

Borax, 732 ; -bead reaction, 733

Bordeaux mixture, 806

Boron, 732 ; adamantine, 736
amide and imide, 738 ; amor
phous, 735 ; bromide, 738
carbides, 736 ; chloride, 738
colloidal, 735 ; crystalline, 736
fluoride, 737 ; halogen com
pounds, 737 ; hydrides, 737
iodide, 738; nitride, 736
oxides, 732, 734, 736, 739
phosphate, 735 ; preparation of,
735 ; sulphate, 735 ; sulphide,
736 ; sub-group, 890

INDEX

1047

Borocalcite, 732

Borofluorides, 737

Borohydrate, 739

Boronatrocalcite, 732

Bort, 660

Boyle, 27, 31, 36, 344, 832 ; law
of, 66

Bragg, researches on crystals and
X-rays of, 1018

Brass, 810, 859

Braunite, 961

Bredig's method for colloidal
silver, 824

Bricks, 744, 901, 902, 947

Brin's process, 169

Brine, 219

Britannia metal, 915, 934

Broggerite, 604

Bromates, 402

Bromide ion, 400

Bromides, 400

Bromine, 393 ; atomic weight of,
396 ; hydrate, 396 ; manufacture
of, 394 ; occurrence, 393 ; pre-
paration of, 393 ; properties of,
395 ; -water, 396

Bronze, 810, 973

Brookite, 929

Brownian movement, 311

Brucite, 857

Brunswick green, 814

Buddling, 913

Burt and Edgar, volume com-
position of water, 215

Cacodyl, 650

Cadmia, 859, 866

Cadmium, 866 ; -compounds, 866

Caesium, 770, 797

Cairngorm, 741

Calamine, 860

Calciner, 645

Calcite, 839

Calcium, 839, 845 ; bromide, 845 ;
carbide, 84T; carbonate, 839;
chlorate, 385 ; chloride, 844 ;
chromate, 947 ; cyanamide, 541,
544, 848 ; fluoride, 845 ; group,
911 hydride, 183, 846; hydro-
sulphide, 847 ; hydroxide, 841 ;
hypochlorite, 380 ; hypophos-
phite, 642 ; iodide, 848 ; man-
ganite, 241 ; nitrate, 563, 848 ;
nitride, 846 ; oxalate, 849 ;
oxide, 840 ; permanganate, 969 ;
peroxide, 843 ; phosphates, 609,
848 ; phosphide, 621 ; plumbate,

924 ; polysulphides, 847 ; sub-
chloride, 845 ; sulphate, 846 ;
sulphide, 847 ; sulphite, 495,
847 ; thiosulphate, 847 ; tung-
state, 957

Calc spar, 840

Caliche, 405

Caloric, 36, 42

Calorie, 201

Calorific power of fuel, 671, 683,
706

Calx, 36

Cannizzaro, 143

Carbides, 671

Carbon, 658 ; allotropic forms of,
659 ; amorphous, 665 ; atomic
weigh£--of,"692 ; combustion of,
700 ; compounds of, 658 ; cycle,
694; dioxide, 686, 698; di-
sulphide, 710 ; equivalent of,
692 ; gas-, 683 ; monosulphide,
713 ; monoxide, 699 ; oxides,
686 ; oxysulphide, 708 ; sub-
oxide, 710 ; subsulphide, 713 ;
sulphoselenide, 714 ; sulpho-
telluride, 714 ; tetrachloride, 712

Carbonado, 660

Carbonates, 690

Carbonyl, 518 ; bromide, 708 ;
chloride, 708 ; sulphide, 709

Carbonyls, 703

Carborundum, 753

Carboxyhasmoglobin, 703

Carbyl sulphate, 677

Carnallite, 791, 856

Carnelian, 743

Carnotite, 958

Case-hardening, 983

Cassel yellow, 925

Cassiterite, 912

Cast iron, 977

Castner-Kellner cell, 296

Catalysis, 166, 198, 225 ; negative,
167, 494 ; photochemical, 234,
695

Catalysts, 164, 166

Catalytic combustion, 198

Cataphoresis, 12, 888

Cathode, 278

Cations, 278

Cat's eye, 744

Caustic potash, 792 ; soda, 779

Caustification, theory of, 779

Cavendish, 35 ; on equivalents,
117; on inflammable air, 180;
on water, 51, 213

1048

INDEX

Gawk, 850

Celestine, 851

Cells, concentration, 887 ; Daniell,
883 ; reversible, 884 ; voltaic,
881 ; voltage of, 884

Cement, 843

Cementite, 978, 983

Ceramics, 901, 903

Cerite, 907

Cerium, 906 ; compounds, 908

Cerussite, 920

Chalcedony, 743

Chalcocite, 805

Chalcopyrite, 805

Chalk, 694, 840

Chalkos, 805

Chalybite, 974

Changes, chemical and physical,
18 ; of state, 106, 269

Charcoal, 665

Charles's law, 67

Charleston phosphate, 608

Chemical changes, 18 ; energy,
387 ; notation and nomenclature,
1 30 ; photometer, 874

Chemiluminescence, 719

Chemistry, early history of, 27

Chessylite, 805, 814

Chile nitre, ftee Sodium nitrate

China, 903

Chlorapatite, 008

Chlorargyrite, 819

Chlorates, 369 ; manufacture of,
385

Chlorine, 218, 220 ; atomic weight
of, 148, 233 ; available, 378 ;
dioxide, 380 ; heptoxide, 385 ;
hydrate, 228; liquid and solid,
226 ; manufacture, 238, 296; mon-
oxide, 372 ; oxygen compounds,
373, 391 ; properties of, 225 ;
water, 228, 369

Chlorites, 372, 381

Chlorochromates, 956

Chlorophosphamide, 624

Chlorophyll, 694, 855

Chlorostannates, 917

Chromates, 954

Chrome, alum, 162, 952 ; -green,
950 ; -ironstone, 947 ; -ochre,
947 ; -tanning, 952 ; -yellow,
927

Chromic compounds, 950

Chromite, 947, 951

Chromitite, 947

Chromium, 947 ; compounds, 953

Chromous compounds, 949

Chromylamine, 956 ; -chloride, 955

Chrysoberyl, 891

Chrysocolla, 815

Chrysoprase, 743

Cinnabar, 868

Cis-isomers, 1013

Clarain, 669

Clark's process, 208

Claudetite, 651

Clausthalite, 528

Clay, 890, 901 ; -ironstone, 974

Cleveite, 604

Coal, 669, 694 ; -gas, 680

Coarse metal, 806

Cobalt, 998; aluminate, 897; -bloom,
644, 998 ; compounds, 998-1001 ;
tin-white, 644

Cobaltammines, 1001 ; constitu-
tion of 1011

Cobaltite, 644, 998

Colcathar, 989

Coke, 683 ; -ovens, 684

Colemanite, 732

Collargol, 824

Collodion films, 315

Colloidal solutions, 8. 12, 314, 888

Colloids, 888 ; dialysis of, 314 ;
diffusion of, 314, 316 ; molecular
weight of, 315 ; osmotic pressure
of, 316

Columbium, see Niobium

Combination form, 434

Combining, capacity, 245 ; -weight,
119

Combustion, 164; fractional, 674;
preferential, 725

Combustion, Theory of : HookeV 36 ;
Lavoisier's, 48 ; Priestley's, 46 ;
Scheele's, 42 ; Stahl's, 38

" Compo ' tubing, 291

Components, 106

Compounds, 23 ; complex, 360,
804, 972, 1012 ; endothermic
and exothermic, 390 ; formula? of,
133, 146 ; metallic, 768 ; meta-
meric, 496 ; molecular, 252 ;
molecular heat of, 431 ; organo-
metallic, 464, 517 ; saturated
and unsaturated, 250 ; stability
of, 389

Compressibility, coefficient, 148 ;
of elements, 456 ; of gases, 66

Concentrated soda crystals, 784

Concentration, 66, 98, 99, 105, 309

Conductivity, equivalent, 289

Constantan, 1004

Contact action, 166

INDEX

1049

Copper, 804; alloys of, 809;

colloidal, 811 ; estimation of,

817; -extraction, 806, 808;

-nitroxyl, 592 ; ores of, 804 ;

oxychlorides, 814 ; peroxides,

815 ; refining, 808 ; silicates,

815 ; sflicide, 815 ; suboxide,.

818; -zinc couple, 182
Goprolites, 608
Coral, 694
Coronium, 468

Corrosive sublimate, 868, 873
Corubin, 947
Corundum, 894
Cottrell process, 16, 508
Coulomb, 279, 880
Coulometer, 57 ; copper, 817 ;

silver, 825
Covelline, 805
Cowper stove, 976
Cracking of oils, 672
Crocoisite, 947
Crocus, 989
Crookesite, 528, 904
Crops, 696
Crucibles, 664
Cryohydrate, 104
Cryolite, 891, 898
Crystal, -axes, 438 ; -carbonate,

784 ; -faces, 439 ; overgrowth-,

447 ; -systems, 436
Crystalloids, 314
Crystals, 1, 433, 1018; mixed-,

446, 886 ; symmetry of, 434 ;

twin-, 442
Cube, 434
Cupel, 819, 822
Cupellation process, 819
Cupric, arsem'te, 653 ; bromide,

814 ; carboriates, 814 ; chloride,
813 ; cyanide, 715 ; halogen
compounds, 813 ; hydroxide,
814; ion. 811; nitrate, 813;
oxide, 776, 811 ; phosphate,

815 ; phosphide, 620, 815 ; sul-
phate, 612 ; sulphide, 813

Cuprite, 805

Cuprous, acetylide, 678, 816 ; chlor-
ide, 816 ; cyanide, 715, 817 ;
hydride, 817 ; iodide, 817 ; ion,
811 ; nitride, 818; oxide, 811 ;
phosphide, 815 ; sulphate, 812,
817; sulphide, 817; sulphite,
817; thiocyanate, 817

Curie, Mme.y 1020

Cyanamide, 544, 848 ; process,
see Ammonia

Cyanates, 718

Cyanides, 682, 717; ion of, 616;

process for gold, 833 ; for silver,

821 ; tests for, 718
Cyanite, 746
Cyanogen, 715; bromide, 718;

chloride, 717 ;Niodide, 718
Cyanuric chloride, 717
Cyclic reactions, 167

Dale and Milner process for white

lead, 929
Dalton, 126

Davy, on chlorine, 221 ; on flame,

723 ; on isolation of alkali

metals, 774

Deacon process, 238, 242
Decomposition, 154 ; double, 154
Degrees of freedom, 92
Deliquescence, 306
Delta metal, 810
Denitrifying bacteria, 576
Density of a gas, limiting, 147 ;

moist, 80 ; normal, 68 ; relative,

68, 142
Density of a vapour, 81 ; Dumas'

method, 83 ; Hofmann's method,

81 ; Nernst's method, 88 ; Victor

Meyer's method, 86
Dephlogisticated air, 44
Detinning process, 915
Detonation wave, 729
Development, 831"
Dewar vessel, 174
Dialogite, 961
Dialysed iron, 989
Dialysis, 170, 314
Diamond, 660
Diaspore, 890
Diborates, 734
Dichloramine, 556
Dichromates, 954
Diffusion, of gases, 14, 191 ; of

liquids, 258, 313
Dihydrol, 201
Dihydroxylamine, 542
Di-imide, 541
Dimorphism, 443, 447
Dioptase, 815
Dioxides, 134, 342
Disilane, 749
Disiloxane, 749
Displacement, chemical, 154
Dissociation, curves, 155 ; degree

of, 152 ; electrolytic, 283, 357 ;

by heat, 151, 349, 354

1050

INDEX

Distillation, 93; fractional, 94;

isothermal, 307 ; under reduced

pressure, 336
Dobereiner's lamp, 198 ; law of

triads, 451
Dolomite, 840, 855
Domes, 439 ; macro- and brachy-,

439

Draper effect, 235
Drier, for paints, 964
Driped, 952
Dry cell, 966
Dulong and Petit's law, 140, 254,

425 ; exceptions to, 427
Dumas, composition of water, 61
Durain, 669
Duralumin, 893
Dutch liquid, 677; metal, 810;

process, 928 ; white, 928

Earth, age of, 1034 ; composition
of, 32, 838

Earthenware, 901

Earths, rare, 461

Eau de Javelle, 368

Efflorescence, 203

Effusion, 263

Einstein's theory of specific heats,
312

Eka-elements, 470, 1029

Eldred's wire, 1007

Electric calamine, 860 ; current,
881 ; furnace, 892, 982 ; lamp,
958; lamp -filament, 930

Electrical pressure, 880 ; work,
880

Electrode potentials, 884

Electrodes, 278; carbon, 684

Electrolysis, theory of, 280 ; laws
of, 277 ; of water, 56

Electrolytes, 278, 287, 292, 316

Electrolytic gas, 57

Electromagnetic separation, 10

Electromotive force, 880

Electrons, 281, 881, 1035 ; orbital,
1036 ; negative, 1016

Electroplating, 825

Electroscope, 1021

Electrotyping, 809

Elements, 23, 32, 1026, 1034;
atomic numbers of, 452, 1031 ;
average life, 1027 ; classification
of, 450 ; compressibility of,
456 ; electrochemical character,
252, 450, 455, 464, 91 1 ; fusi-
bility, 455 ; half -life, 1027 ; inac-
tive, 598 ; isomorphous, 443 ;

melting- and boiling-points, 458 ;
molecular weight of, 147 ; names
of, 131 ; oxygen compounds of,
463 ; rare earth-, 461 ; symbols
of, 132 ; theory of four, 27 ;
transitional, 460 ; transmutation
of, 1027 ; volatility of, 455

Emanation, of radium, 1025 ; of
thorium, 1028

Emerald, Oriental, 894 ; Peruvian,
854

Emery, 894

Emulsion, 14

Enantiomorphism, 742

Energy, chemical, 879, 924 ; con-
servation of, 388 ; free, 879 ;
total, 879 ; -quanta, 429

Enstatite, 746

Equations, chemical, 136

Equilibrium, 76 ; chemical, 152,
183, 344, 347; constant, 347,
352 ; effect of temperature and
pressure on, 355 ; effect of
products of reaction on, 356 ;
kinetic, 270

Equivalent weight, 119 ; Cavendish
on, 117; of element, 246;
standard of, 123

Erubescite, 805

Estramadurite, 608

Estrich-gips, 846

Ethane, 677

Ethyl borate, 738 ; hyponitrite,
594 ; nitrite, 587 ; orthosilicate,
748 ; peroxide, 336, 341 ; phos-
phite, 640

Ethylene, 667, 675 ; dibromide, 677 ;
dichloride, 677

Eudiometer, 52, 58

Eutectic, 104 ; point, 767

Euxenite, 907

Evaporation, 270 ; latent heat of,
304 ; in vacuum, 780

Expansion, adiabatic, 171, 599 ;
coefficient of, of gases, 67

Explosion, of electrolytic gas, 51 ;
of gunpowder, 565 ; of hydrogen
and chlorine, 234

Explosion wave, 729

Explosives, 801

Faience, 903

Fajans and Soddy's law, 1032

Far dflai/t 279; laws of electrolysis,

277, 279
Fats, hardening of, 191, 1005

INDEX

1051

Fehling's solution, 815

Felspar, 789, 891

Fergusonite, 907

Fermentation, 686

Ferrates, 987

Ferric bromide 991 ; chloride,

990 ; fluoride, 991 ; hydroxide,

989 ; ion, 986 ; nitrate, 991 ;

oxide, 986, 989 ; phosphate,
. 991 ; salts, 987 ; sulphate, 991 ;

sulphide, 992 ; thiocyanate, 996
a-ferric ferricyanide, 996
a -ferric ferrocyanide, 995
Ferrites, 987
Ferrochrome, 958 ; -manganese,

961 ; -molybdenum, 957 ; tung- Fusain, 669

Fracture, crystalline and coiichoidal,
2, 433

Franklinite, 860

Fraunhofer lines, 760

Freezing machines, 547 ; point,
depression of, 103 ; do., abnor-
mal, 317 ; do., molecular, 299

Fremy's salt, 420

Froth, 15

Fumaroles, 542

Fume, 15

Funnel, separating, 98

Furnace, electric, 858 ; muffle, 238,
822 ; reverberatory, 920 ; re-
volving, 778

Fusion, latent heat of, 271, 300 ;
-mixture, 792

sten, 958

Ferroso-ferric iodide, 487 ; -hydr-
oxide, 987, 990 ; oxide, 182, 987,
990

Ferrous ammonium sulphate, Gadolinite, 907

988 ; bicarbonate, 988 ; bromide, Gaillard tower, 507

987 ; carbonate, 988 ; chloride, Galena, 819, 920

988 ; chromite, 947 ; hydroxide, Galician potash deposits, 791
988 ; iodide, 987 ; ion, 985 ; Gallium, 470, 890, 904
oxide, 986, 988 ; salts, 985 ; Galvanising, 862

sulphate, 185, 987; sulphide, Gamboge, 311; Perrin's experi-
ments with, 313

992 ; sulphite, 991 ; thiosul
phate, 991 ; titanate, 929 ; tung-
state, 957

Fertilisers, 696

Fibrox, 753

Filter pump, 13

Filtration, 13

Fine solder, 915; metal, 806

Fireclay, 902

Firedamp, 622

Fire extinguishers, 689

Flame, 683, 718; Bunsen, 726,
728 ; hydrocarbon, 722 ; lu-
minosity of, 722, 724 ; structure
of, 721, 727 ; temperature of,
728

Flotation process, 10, 13

Fluorapatite, 416

Fluorescence, 8, 415

Fluorides, 421

Fluorine, 415, 418

Fluorspar, 415, 737, 845

Fluosilicates, 753

Flux, 415

Foam, 15

Fog, 15

Formaldehyde, 674

Fordos and-Gelis' salt, 836

Formula, of compound, 133 ;
empirical, 136 ; structural, 249

Gangue, 415

Garnet, 746

Garnierite, 1002

Gas, 30 ; absorption by charcoal,
667 ; compression of, 66 ; -con-
stant, 149 ; densities, limiting,
147, normal, 68, 72, relative,
68 ; discovery of, 35 ; equation,
149 ; expansion by heat, 67 ;
ionisation of, 1018, 1021 ; kinds
of, 3 ; liquefaction of, 170 ;
natural, 604, 672 ; separation,
667 ; viscosity of, 266

Gay-Lussac's law of volumes,
138 ; tower, 505

Geber, 29

Germanium, 470, 929

German silver, 1004

Glass, 849 ; of antimony, 935; boro-
silicate, 927 ; Bohemian, 850 ;
devitrification of, 850 ; Jena,
850 ; optical, 905 ; ruby, 815,
835

Glauberite, 847

Glauber salt, 229, 513

Glaze, 902

Glover tower, 504

Glucinum, see Beryllium.

Glue, 609

1052

INDEX

Glycerophosphates, 609
Glycol, 677

Gold, 832 ; compounds, 835 ; col-
loidal, 834 ; fulminating, 836 ;
leaf, 834 ; nitride, 836 ; plating,
834 ; standard, 823 ; trichloride,
833

Goldschmidt's thermit process,
894, 948

Gothite, 989

Goulard's extract, 928

Graham, law of diffusion of, 191 ;
on colloids, 314

Gram molecular volume, 149 ;
weight, 149

Graphite, 663, 670

Graohon sulphate, 665

Greenockite, 866

Green vitriol, 987

Groups, negative, 517

Guano, 628

Guignet's green, 951

Guldberg, law of mass action, 353

Gun metal, 810

Gunpowder, 564

Gutzeit's test for arsenic, 650

Haber process, see Ammonia,
synthetic

Haematite, 974

Haemocyanin, 805

Haemoglobin, 697

Halogens, 393, 422

Hargreaves-Bird cell, 296

Hausmannite, 961

Heat, animal, 697 ; of evaporation,
201 ; of formation, 388 ; of
fusion, 201 ; mechanical equiva-
lent of, 201 ; of reaction, 387 ;
regenerators, 977 ; specific, 201

Heavy spar, 850

Helium, 603 ; atom, 1023, 1035 ;
group, 605 ; from radium, 1023,
1026

Helmont, Van, 30

Hemihedral forms, 440 ; of hexa-
gonal system, 441

Henry's Law, 96, 272

Hess's Law, 388

Hessite, 532

Heterogeneous bodies, 6, 26

Heterotypes, 1033

Hexagonal system, 438

Hexakisoctahedron, 436

Holohedral forms, 440

Homogeneous bodies, 6, 26

Homologous series, 672

Hooke, theorv of combustion of, 36

Hornblende/ 890

Horn silver, 819, 827

Humidity, 78

Humus, 696

Hydrargillite, 891

Hydrargyros, 868

Hydrates, 101 ; vapour pressure of,
203

Hydrazine, 541, 557 ; hydrate,
558 ; salts, 557

Hydrides, 188

Hydrocarbons, 672

Hydrogel, 745

Hydrogen, 180 ; atom, 1035 ; com-
bining volume with oxygen, 213 ;
compressibility of, 148 ; density,
72 ; liquid and solid, 192 ;
manufacture of, 183, 707 ; mass
of atom, 128, 268 ; nascent, 189;
occlusion of, by metals, 194 ;
properties of, 187 ; pure, 186 ;
spectrum of, 187 ; union of,
with chlorine, 234 ; uses of, 190

Hydrogen bromide, see Acid,
hydrobromic

Hydrogen chloride, see Acid,
hydrochloric

Hydrogen iodide, see Acid, hydr-
iodic

Hydrogen peroxide, 333 ; estima-
tion of, 339 ; formula of, 341 ;
hydrates of, 337 ; properties of,
336 ; pure, 335

Hydrogen persulphides, 487 ;
phosphides, 618 ; selenide, 530 ;
sulphide, 483 ; telluride, 532

Hydrogenite, 183

Hydrogenmm, 195

Hydrohaematite, 989

Hydrol, 201

Hydrolith, 183, 846

Hydrolysis, 206, 360

Hy drone, 182

Hydrosol, 145

Hydroxylamine, 552 ; compounds,
552

Hygroscopic substances, 307

Hypoantimoniates, 935

Hypoborates, 737

Hypochlorites, 368, 380

Icositetrahedron, 436
Ignition points, 723
Ilmenite, 929
Indicators, 363

INDEX

1053

Indigo blue, 369

Indium, 890, 904 ; atomic weight of,
468

Induction, period of, 405

Inflammable air, 35, 49

Infusible white precipitate, 876

Ink, 986

Iodine, 402; acetate, 411; atomic
weight of, 414 ; chlorides, 410 ;
manufacture of, 404 ; oxides,
411 ; perchlorate, 411 ; penta-
fluoride, 411 ; properties, 400 ;
pure, 406; sulphate, 411 ; tests
for, 407 ; tincture of, 407

lodonium compounds, 411

lodothyrin, 403

Ionic theory, difficulties of, 285

lonisation constant, 358 ; degree of,
291 ; in stages, 295 ; of salts,
table, 318

Ionium, 1028

Ions, 15, 879 ; gaseous, 1021 ;
migration of, 287 ; mobility of,
288 ; nomenclature of, 278,
285 ; osmotic pressure of, 885

Iridium, 1010

Iron, 973 ; allotropic forms of, 982 ;
alums, 991 ; carbonyls, 992 ;
compounds, see Ferric and Fer-
rous salts ; dinitrosothiosul-
phates, 997 ; estimation of,
955, 970 ; malleable, 978 ; metal-
lurgy of, 975 ; native, 974 ;
oxides of, 974, 986 ; passive,
985 ; pure, 983 ; pyrites, 993 ;
rusting of, 983 ; sulphides, 992

Irreversible reaction, 353

Isatin, 369

Isocyanides, 717

Isomeric change, 156

Isomerism, 114; of complex com-
pounds, 1012 ; types of, 1013

Isomorphism, 146, 442, 447 ; ex-
ceptions to law of, 447

Isomorphous mixture, 446

Isotopes, 114, 462, 1033 ; spectra of,
1033

Ivory black, 118

Jasper, 744

Jet, 670

Joule, 880 ; on expansion of gases,
258 ; Law of molecular heat,
431

Joule-Kelvin effect, 172 ; in hydro-
gen, 192

Kainite, 791, 855

Kaolin, 891, 902

Kaolinite, 746, 901

Kelp, 402, 790 ; salt, 404

Kermes mineral, 938

Kieselguhr, 741

Kieserite, 855

Kilowatt, 880 ; -hour, 880

Kinetic theory, of equilibrium,
347 ; of gases, 258 ; of liquids,
270 ; of solids, 271 ; of solution,
271

King's yellow, 655

Kipp's apparatus, 185

Kish, 663

Kobold, 998

Kohlrausch method for conduc-
tivity, 292

Kryptol, 664

Krypton, 603, 605

Kupfer-nickel, 1002

Lake, 896

Lamp black, 668

Lamps, electric, 602

Lanarkite, 920

Landold, on Conservation of Mass,
22

Landsberger's boiling point ap-
paratus, 306

Lapis lazuli, 903

Lavoisier, antiphlogiston theory,
46 ; on air, 48 ; on chlorine, 371 ;
on water, 54

Lead, 920, 1027 ; accumulator, 923 ;
atomic weight, 921 ; acetates.
928 ; borate, 927 ; carbonates,
925 ; chromates, 927, 955 ;
colloidal, 921 ; hydroxide, 922 ;
oxides, 922 ; halogen com-
pounds, 925 ; isotopes of, 462 ;
nitrate, 926 ; phosphates, 927 ;
pyrophoric, 166 ; silicates, 927 ;
sulphates, 926 ; sulphides, 920,
926

Lead, argentiferous, 819

Leadhillite, 920

Leblanc process, 777

Lecithins, 609

Leclanche cell, 966

Lepidolite, 795

Leucite, 746

Leucone, 750

Liebig's condenser, 93

Lignite, 670

1054

INDEX

Lime, burning, 841 ; -kiln, 841 ;

-light, 189; slaked, 842; -stone,

839 ; superphosphate of, 849 ;

-water, 842
Limonite, 974, 989
Linotype metal, 934
Lipowitz' alloy, 941
Liquids, 3, 270
Liquor of flints, 739
Lithium, 770, 795 ; compound's of,

796-797 ; mica, 795
Lithopone, 853
Litmus, 363

Litre, Mohr's, 200 ; standard, 2UO
Liver of sulphur, 795
Loam, 891
Lorandite, 904
Lowig process, 784, 990
Luce-Rozan process, 820
Luminous paint, 877
Luna cornea, 827
Lunar caustic, 826
Lungs, 697
Luzi's test, 664

Mac Arthur and Forrest, process

for gold, 833
Magnalium, 893

Magnesia, 857; -alba, 855; -alba
levis, 858 ; -alba ponderosa,
859 ; calcined, 858 ; fluid,
859 ; -mixture, 859

Magnesite, 855, 858

Magnesium, 855 ; ammonium phos-
phate, 632, 859 ; arsenates, 654 ;
arsenide, 857 ; atomic weight
of, 859 ; boride, 737 ; bromide,

856 ; carbides, 857 ; carbonates,
858; chloride, 856, 857 ; hydr-
oxides, 857; iodide, 856 ; metallic,

857 ; mixture, 654 ; nitride,
857 ; oxides, 855, 857 ; phos-
phate, 632, 859; phosphide,
857; silicide, 748, 857; sul-
phate, 855 ; sulphides, 857

Magnetite, 974
Majolica, 903
Malachite, 805, 814
Manganates, 966

Manganese, 960 ; bo rate, 964 ;
bronze, 961 ; carbide, 961, 964 ;
chlorides, 964 ; cyanogen com-
pounds, 970 ; dioxide, 164, 166,
fluoride, 965 ; heptoxide,
metallic, 961 ; nitride,
ores, 960 ; oxides, 962,

965
968
961

965, 968 ; steel, 961

Manganic compounds, 904, 000

Manganin, 961, 965

Manganite, 961, 964

Manganous compounds, 962-904

Mansfield process, 807

Manures, 696

Marble, 840

Marcasite, 993

Margarine, 1005

Marl, 891

Marsh-Berzelius' test, 648

Marsh gas, 672

Martensite, 983

Mass Action, Law of, 344, 350 ;
applied to ionisation, 357 ; to
reactions, 362

Matches, 626

Matte, 807

Matter, law of conservation of, 19,
128 ; structure of, 125

Mayow, 37, 344

Meerschaum, 855

Meiler, 666

Melaconite, 805

Melilith, 746

Mercaptan, 496, 625

Mercurammonium compounds,
876

Mercuric, acetylide, 875 ; bromide,
875 ; carbonate, 876 ; chloride,
872, 873 ; cyanide, 875 ; fluor-
ide, 875 ; fulminate, 876 ;
iodide, 116, 875; nitrate, 872;
nitride, 872 ; oxide, 872 ; sul-
phate, 872 ; sulphide, 868, 876 ;
thiocyanate, 876

Mercurius prascipitatus per se,
873

Mercurous, compounds, 871, 872 ;
iodide, 116 ; ion, 870

Mercury, 764, 867 ; amalgams,
870 ; compounds of, 870 ; col-
loidal, 869 ; metallurgy of, 868 ;
oxychlorides, 874 ; periodide,
875 ; peroxide, 873 ; properties
of, 869 ; purification of, 869

Mesothorium, 1028

Meta- alumina tes, 896 ; -aluminium
hydroxide, 896 ; -borates, 734 ;
-elements, 906 ; -phosphates,
633 ; -phosphoryl chloride, 637

Metalloids, 931

Metals, 764 ; allotropic forms of,
765 ; combustion of, 35 ; cal-
cination of, 35 ; electromotive
series of, 886; noble, 818;
single potentials of, 885; pro-

INDEX

1055

perties of, 451, 764 ; solution
pressure of, 884 ; welding of,
189

Metamerism, 496

Metamers, 156

Meteorites, 33, 974

Methane, 672, 713

Methyl-orange, 364 ; -red, 363 ;
-violet, 364

Mica, lithium-, 795 ; potash-, 746

Microbalance, 88

Microcosmic salt, 542, 608, 631

Microlith, 907

Microns, 8

Mild steel, 982

Milk, of lime, 842 ; potassium in,
790

Millerite, 746, 1002

Millon's base, 877

Mimetisite, 644

Mineral chameleon, 967

Minerals, formulae of, 446

Minium, 923, 924

Mirrors, 824, 915

Mispickel, 644

Mist, 15

Mixed crystals, 446

Mixtures, isomorphous, 446 ; me-
chanical, 7 ; separation of, 9

Moebius' electrolytic process for
silver, 822

Moh's scale of hardness, 661

Mohr's salt, 988

Moissan on diamond, 662 ; on
fluorine, 417

Moist gases, 76 ; density of, 80

Moisture, catalytic effect of, 704

Molecular, magnitudes, 266, 268 ;
-weight, 142 ; -weights of col-
loids, 315 ; -weights by freezing
point, 301 ; -weights in solution,
299, 317

Molecules, attraction of, 269;
Avogadro on, 138 ; diameter of,
264 ; energy of, 262 ; existence
of, 268 ; gaseous, 1 40 ; motion
of, 258 ; speed of, 258, 259, 262

Molybdenite, 957

Molybdenum, 957 ; compounds of,
957

Molybdoena, 663

Monazite, 930

Mond, carbonyl process, 1002 ; -gas,
706

Monel metal, 1002

Monochloramine, 556

Monoclinic system, 440

Monosilane, 748

Monotype metal, 934

Mordants, 895

Morley, density of a gas, 71 ; on

water (composition of), 63
Mortar, 843
Mosaic gold, 919
Moseley, 1030
Muffle furnace, cupellation, 822 ;

salt cake, 238
Muntz metal, 810
Muscovite, 789

Naples yellow, 925

Nebuhe, 33

Nebulium, 33, 468

Neodymia, 906

Neon, 603, 604

Nernst lamps, 929

Nessler's reagent, 875

Neutralisation, 294 ; heat of, 295

Niccolite, 1002

Nichrom, 1004

Nickel, 1002 ; alloys, 1004 ; cata-
lytic action of, 1005 ; com-
pounds of, 1003 ; dimethyl
glyoxime, 1005; estimation of,
1005 ; glance, 644, 1002 ; metal-
lurgy of, 1002 ; ochre, 1002 ;
plating, 1003

Niobium, 944

Niton, 605, 1025

Nitrates, estimation of, 579 ; manu-
facture of, 574, 576 ; occurrence
of, 563

Nitre, 563 ; -air, 36

Nitric oxide, 578 ; composition of,
580 ; properties, 579

Nitrifying bacteria, 563, 576

Nitrites, 584 ; estimation of, 579

Nitrobenzene, 569

Nitroethane, 587

Nitrogen, 535, 607 ; active, 541 ;
atmospheric, 535, 537 ; atomic
weight of, 550 ; combined, 536 ;
compounds with hydrogen,
541 ; cycle, 576, 577 ; density
of, 600 ; dioxide, 590 ;
fixation of, 544, 848 ; group,
607, 931 ; iodide, 556 ; manu-
facture of, 538 ; oxides, 561 ;
oxy-acids, 561 ; pentoxide, 577 ;
preparation of, 536, 538 ; pro-
perties of, 540 ; sulphides, 596,
597; tetroxide, 590; tribromide,
556 ; trichloride, 539, 554

1056

INDEX

Nitroglycerin, 570
Nitrometer, 579
Nitroso-group, 505
Nitrosulphuryl chloride, 596
Nitrosyl, bromide, 595 ; chloride,

594 ; fluoride, 595 ; perchlorate,

595 ; sulphate, 595

Nitrous anhydride, 58.1, 587 ; oxide,

581-583
Nitroxyls, 592
Non-metals, 450; electromotive

series of, 887
Norgine, 405
Normal solution, 970
Nuggets, 833

Occlusion of hydrogen in palla-
dium, 196

Octahedron, 434, 436

Ohm's Law, 293

Oleum, see Sulphuric Acid,
fuming

Olivine, 746. 855

Onofrite, 528

Onyx, 743

Opal, 743

Open-hearth process, 981

Optical activity, 1015

Organo-metallic compounds, 254

Orichalcum, 859

Orpiment, 644

Orthite, 907

Orthoclase, 746

Osmiridium, 1006, 1010

Osmium, 1010

Osmotic pressure, 307 ; of col
loidal solutions, 316

Ostwald's dilution law, 357 ;
theory of indicators, 363

Overgrowth crystals, 447

Oxidation and reduction, 255,
285, 888

Oxides, acidkTand basic, 776; types
of, 135

Oximes, 554

Oxy-acetylene blowpipe, 188

Oxygen, 24, 43, 159 ; absorption of,
166 ; atomic weight of, 148 ;
combining volume with hydro-
gen, 213 ; compounds of, 246 ;
compressibility of, 148 ; density
of, 72 ; liquid, 175 ; manu-
facture of, 168 ; properties of,
(chemical) 164, (physical) 178

Oxy-hydrogen blowpipe, 188

Oxy-muriatic acid, 221

Oxy-nitrososulphuric anhydride,

596
Ozone, 320, 617 ; density of, 326 ;

formula of, 323, 329 ; liquid,

328 ; manufacture of, 331 ;

properties, 328 ; stability of,

328 ; tests for, 330
Ozonic acid, 330
Ozonides, 329

Palladium, 186, 197, 1009 ; hydride,
195

Paper, 846

Papyrus of Ley den, 28

Paracelsus, 29, 180

Paracyanogen, 715

Paraffin, 671

Parkes process, 820

Passivity of metals, 949

Pattinson process, 820

Pearl, ash, 789, 791 ; -hardening
846 ; -white 940

Pearlite, 983

Peat, 669

Peligot's salt, 956

Pencils, black lead, 664

Pentlandite, 1002

Perborates, 738

Percarbonates, 693

Perchloric anhydride, 385

Perhydrol, 334

Periclase, 857

Periodates, 414

Periodic, law, 453, 456 ; system,
450, 459, 462, 471 ; table, 450,
466, 1030

Permanent white, 853

Permanganates, 960, 966

Permanganyl chloride, 968 ; fluor-
ide, 968

Peroxides, 134, 342, 838

Perrin, Jean, 311

Per sulphates, 519

Perthiocarbonates, 715

Petalite, 746, 795

Petrifaction, 741

Petrol, 671

Petroleum, 671

Pettenkofer's method, 699

Pewter, 915, 934

Pharaoh's serpent, 876

Pharmacolite, 644

Phase Rule, 106

Phases, 7

Phenol, 511

Phenol phthalein, 363

Philosopher's stone, 29 ; wool, 862

INDEX

1057

Phlogisticated air, 46, 535

Phlogiston, theory of, 38

Phosgene, 708

Phospham, 624

Phosphamide, 625

Phosphates, 630

Phosphine, 618

Phosphonitrile chlorides, 625

Phosphonium , bromide, 620; chlor-
ide, 620 ; iodide, 620

Phosphor-bronze, 810, 915; -tin, 915

Phosphorescence, 611, 614, 877,
1021

Phosphoretted hydrogen, gaseous,
618; liquid, 618, 621; solid,
618, 622

Phosphorus, 607, 608 ; allotropic
forms of, 614 ; amorphous, 612 ;
Baldwin's, 848 ; burning of,
626 ; dichloride, 625 ; glow of,
614 ; halogen compounds of,
622 ; iodides, 625 ; oxides, 626 ;
oxy-acids, 626 ; oxy-bromide,
635; oxy-chloride, 624; oxy-
fluoride, 635 ; pentabromide, 151,

625 ; pentachloride, 623 ; penta-
fluoride, 623 ; pentoxide, 627 ;
preparation of, 609 ; purification
.of, 610 ; red, 612 ; suboxide,

626 ; sulphides, 625 ; tetroxide,
638 ; tribromide, 625 ; tri-
chloride, 623 ; trifluoride, 623 ;
trioxide, 637 ; uses of, 610 ;
white, 611

Phosphoryl chloride, 624 ; radical,
635

Photochemical induction, 235

Photography, 830

Photophone, 529

Photosensitisers, 830

Photosynthesis, 894

Pig iron, 977, 978

Pinakoids, 439 ; basal-, 439 ;
brachy-, 439 ; macro-, 439

Pintsch gas, 707

Pitchblende, 958, 1021

Planck's constant, 429

Plants, growth of, 696

Plaster of Paris, 846

Plateau's soap solution, 14

Platinic chloride, 1006

Platinised asbestos, 1008

Platinoid, 1004

Platinum, 1006, 1007 ; -black, 1007 ;
catalytic action of, 198 ; col-
loidal, 1007 ; compounds of,
1007, 1009 ; -sponge, 1007

Plattner's chlorine process for
gold, 833

Plattnerite, 920

Plumbago, 663

Plumbates, 924

Plumbic chloride, 925 ; sulphate,
927

Plumbum candidum, 912 ; cine-
reum, 940 ; nigrum, 912

Pneumatolysis, 694

Polarisation, electromotive force
of, 292

Polarised light, 1014

Polymerism, 327

Polymers, 156

Polymorphism, 443, 447

Polyphosphides, 622

Porcelain, 901, 902

Portland cement, 843

Positive nucleus, 1035

Positive rays, 1016

Potash, 772, 789, 792

Potassamide, 547

Potassium, 770, 775, 794 ; amino-
chromate, 956 ; antimonyl tar-
trate, 940 ; argent o cyanide, 360,
825 ; auro cyanide, 833 ; auri-
cyanide, 836 ; borates, 793 ;
bromate, 401 ; bromide, 792 ;
carbonates, 789, 791, 792 ;
chlorate, 370, 386 ; chloraurite,
835 ; chloride, 792 ; chloro-
chromates, 956 ; chromium sul-
phate, 952 ; cobaltinitrite, 999,
1001 ; cobaltocyanide, 1001 ;
cuprocyanide, 817 ; cyanate,
716, 793 ; cyanide, 716, 793 ;
dichromate, 947, 954 ; ferrate,
973, 993 ; ferricyanide, 994 ;
ferrisulphate, 992 ; ferrite, 993 ;
ferrocyanide, 993 ; ferrous ferro-
cyanide, 995 ; fluorides, 792 ;
hydride, 794 ; hydrogen fluoride,
420 ; hydroxide, 792 ; iodate,
413 ; iodide, 792 ; mangani-
cyanide, 970 ; manganocyanide,
970 ; mercurinitrite, 877 ; ni-
trate, 563 ; nitrite, 585 ; nitroso-
hydroxylaminosulphonate, 557 ;
oxalate, 789 ; oxides, 794 ; per-
carbonate, 693 ; perchlorate, 161,
372, 384 ; periodate, 414 ; per-
manganate, 360, 967, 968 ;
persulphate, 519 ; phosphate,
793 ; phosphide, 622, 793 ;
plumbate, 924 ; plumbite, 922 ;
properties, radio-active, 790 ;

3 Y

1058

INDEX

salt deposits, 790 ; salts, reagent
for, 1001 ; seleno cyanide, 528 ;
selenosulphate, 531 ; sulphates,
512, 513 ; tartrates, 789 ; thio-
cyanate, 793 ; xanthate, 714

Powder of Algaroth, 936

Praseodymia, 906

Precipitation, 358 ; electrostatic, 508

Precht's process, 791

Pressure, critical, 171 ; gaseous,
258, 260 ; partial, 72

Priestley, 44

Priorite, 907

Producer gas, 705

Proportions, Constant, law of, 110,
129; Equivalent, law of, 117.
129 ; Multiple, law of, 115, 129

Protyle, 465

Prussian blue, 994, 995

Pseudomorph, 478

Psilomelane, 961

Puddling process, 978

Pure substances, 5, 26, 112, 1033

Purple of Gassius, 835

Pyrargyrite, 819

Pyrex, 713

Pyrites cinders, 974 ; copper-,
805 ; iron-, 993

Pyrographitic oxide, 664

Pyrolustite, 960, 965

Pyromorphite, 920, 927

Pyrophosphates, 632

Pyrophosphoryl chloride, 637

Pyrosulphuryl chloride, 516

Quantum theory, 429

Quartation, 834

Quartz, 740, 741 ; glass, 743 ;

optical properties of, 742
Quicklime, 840
Quicksilver, 867
Quintessence, 28

Radicals, 136 ; valency of, 251

Radioactive changes, 1026 ; equi-
librium, 1026

Radioactivity, 1020

Radio-elements, 1016, 1032

Radiothorium, 1028

Radium, 958, 1021 ; compounds of,
1022 ; emanation, 88, 1025 ;
metallic, 1022

Rain, 15

Raphides, 849

Rare earths, 906 ; minerals of,
907 ; separation of, 907

Rays, actinic, 695; alpha-, 266,
1022 ; beta-, 1022 ; gamma-,
1022 ; infra-red, 755 ; ultra-
violet, 695, 755

Reactions, condensation-, 517, 522 ;
law of, 355 ; reversible, 183, 346 :
successive, 405

Realgar, 644

Recalescence, 983

Redonda phosphate, 609

Red prussiate of potash, 994

Reducing and oxidising agents,
586

Reduction, 60, 188, 255

Regelation, 91

Reinsch's test, 652

Resin of copper, 816

Respiration, 686, 697

Retarders, 831

Reverberatory furnace, 806

Rey, Jean, 36

Rheostan, 1004

Rhodium, 1010

Rhodonite, 961

Rhombdodecahedron, 436

Rhombic system, 438

Rhombohedron, 441

Rinman's green, 863, 1000

Rio Tinto process, 808

Rocks, primary, 789 ; silicate. 739

Roman cement, 843

Rose process, 834

Rose's metal, 941

Rouge, 989

Roussin's salts, 997

Rubber, molecular weight of, 316;
vulcanisation of, 477

Rubidium, 797

Ruby, 894 ; artificial, 894

Ruthenium, 1010

Rutile, 929

Saccharates, 852

Safety lamp, 723

Sal ammoniac, 799

Sal sedativum, 732

Sal volatile, 801

Salts, 134, 274, 771, 776 ; complex,
856 ; double, 856 : of lemon,
789; Schlippe's, 938; of sorrel,
789 ; of tartar, 789

Saltcake, 777 ; -process, 238

Sand, 741

Sandstone, 741

Sapphire, 894

Sard, 743

INDEX

1059

Sardonyx, 743

Satin spar, 846

Scale, boiler, 207

Scalenohedron, 441

Scandium, 470

Scheele, on chlorine, 221; on fire

and air, 39

Scheele's green, 653, 813
Scheelite, 957
Schlempe, 790
Schumann rays, 759
Schweinfiirter green, 653
Schweitzer's reagent, 818
Scotch hearth, 920
Sea, 790

Sedimentation, fractional, 11
Segar cones, 902
Seggars, 901
Selenite, 846
Selenium, 528 ; compounds of,

528-530 : test for, 531
Semi-metals, 864
Semi-permeable membrane, 307
Senarmonite, 934
Sensitisers, 831
Separating funnel, 14
Serpentine, 746
Shank's lixiviating tanks, 778
Sheffield plate, 825
Sherardising. 863
Siderite, 974, 988
Sidot's blende, 865
Siemens ozoniser, 322 ; -Martin

process, 981
Silica, 253, 740 ; amorphous, 743 ;

colloidal, 745 ; detection and

estimation, 744 ; gelatinous, 744,

752 ; pure, 744
Silicates, 745 ; rocks, 890
Silicium, 740
Silicofluorides, 753
Silicoformic anhydride, 750
Silicon, 739, 747; -acetylene, 749;

adamantine, 748 ; amorphous,
747 ; borides, 753 ; bromides,
750 ; bromoform, 751 ; carbide,

753 ; chlorides, 749 ; chloro-
form, 750 ; disulphide, 753 ;
-ethane, 749 ; fluoride, 751 ;
fluoroform, 751 ; graphitoidal,
747 ; halogen compounds, 749 ;
hydrides, 748 ; iodides, 750 ;
iodoform, 751 ; nitrides, 753 ;
octachloride, 750 ; octahedral,
747 ; organic compounds of,

754 ; oxychloride, 750
Silicone, 749

Silk, artificial, 818

Sillimanite, 901

Siloxicon, 753

Silver, 818, 823 ; acetylide, 830 ;
alloys, 822 ; antimonide, 939 ;
arsenate, 655, 830 ; arsenide,
649 ; arsenite, 653, 830 ; azide,
558 ; bromate, 401 ; bromide,

828 ; carbonate, 827 ; chlorate,

829 ; chloride, 823, 827 ; chro-
mate, 828, 955 ; disulphide,
829 ; dithionate, 829 ; electro-
plating with, 825 ; estimation,
828 ; ferricyanide, 995 ; fluor-
ide, 827 ; fulminating, 827 ;
hydroxide, 827 ; hypochlorite,
829 ; hyponitrite, 593 ; hypo-
phosphate, 641 ; iodide, 828 ;
ion, 826 ; metallurgy of, 819 ;
native, 819 ; nitrate, 826 ; ni-
trite, 585, 826 ; ores, 819 ;
oxides, 827 ; permanganate, 967 ;
phosphates, 633, 829 ; phos-
phide, 830 ; sulphates, 829 ;
sulphide, 818, 829 ; sulphite,
829 ; thiocyanate, 828 ; thio-
sulphate, 828

Sinter, 741

Slags, 739; basic, 628, 981 ; Thomas,
981

Slate, 890

Smalt, 998

Smaltite, 998, 1002

Smithells's experiments on
flame, 727

Smithsonite, 860

Smoke, 15

Soap, 206, 781 ; solution, 14

Soda, 772 ; -ash, 783 ; caustic, 779;
crystals, 784 ; -lime, 542

Sodamide, 547

Sodium, 777, 786 ; alloy with lead,
182 ; aluminate, 892 ; amal-
gam, 182 ; analysis, 788 ; argen-
tocyanide, 822 ; arsenide, 648 ;
arsenite, 653 ; aurosulphide,
836 ; bicarbonate, 782, 785 ;
bismuth thiosulphate, 943 ;
borate, 734 ; carbonate, 784 ;
chlorate, 386 ; chloride, 218,
359 ; chromate, 947, 954 ; cyan-
amide, 788 ; cyanide, 788 ;
ferricyanide, 994 ; ferrite, 784,
989 ; formate, 709 ; hydride,
788 ; hydroxide, 779 ; hydrogen
peroxide, 788 ; hydrosulphide,
795 ; hydro xylamine sulpho-

1060

INDEX

nates, 553 ; hypochlorite, 368 ;
hyponitrite, 593 ; hypophos-
phite, 642 ; hyposulphite, 522 ;
iodate, 405 ; metabisulphite,
495 ; nitrate, 563 ; nitrite, 585 ;
nitroprusside, 996 ; nitrosohy-
droxylamine sulphonate, 580 ;
oxalate, 710 ; oxides, 787 ; per-
carbonate, 693 ; peroxide, 334,
787 ; phosphates, 631, 633 ;
sulphates, 498, 513, 520, 777;
sulphides, 795 ; sulphite, 495 ;
tetrathionate, 522 ; thioanti-
moniate, 938 ; thiocarbonate,
714 ; tungstate, 957 ; stannate,
918 ; stannite, 916

Sodyl hydroxide, 788

Soil, 789

Solder, fusible, 941 ; soft, 915

Solids, 1, 271 ; vapour pressure of,
76

Solubility, 2 ; determination of,
96, 102 ; -product, 358, 780

Solution, of gas, 95, 271 ; heats of,
389;' pressure (electrolytic), 884 :
theory of (Arrhenius'), 283 ;
theory of (gaseous), 311 ; theory
of (hydrate), 202

Solutions, 25, 93 ; boiling point of,
303 ; conductivity of. 289 ;
colloidal, 8, 95, 315, 745; of
gases in liquids, 95 ; of liquids
in liquids, 98 ; of solids in
liquids, 99 ; freezing points of,
103, 299 ; molecular weights in,
299 ; osmotic pressure of, 307 ;
solid, 94, 196, 446 ; vapour
pressure of, 104, 302

Sombrerite, 609

Soret on ozone, 324

Sorption, 197

Sound, velocity of, 263, 599

Spathic iron ore, 974, 988

Specific heats of gases, 146, 598 ;
of solids, 146

Spectra absorption, 762 ; of gases,
759 ; of liquids, 759 ; of solids,

759 ; phosphorescence, 907 ; of
stars, 33 ; variation of, 759

Spectroscope, 757

Spectrum, analysis, 756 ; -bands,
757 ; continuous, 755 ; infra-
red, 759 ; lines, 756 ; solar-,

760 ; ultra-violet, 759
Speculum metal, 810
Speiss-cobalt, 998
Spelter, 860

Spiegel, 961, 979

Spinel, 891, 897

Spinthariscope, 267, 1023

Spodumene, 796

Stahl, 38

Stalactites, 207

Stalagmites, 207

Standard temperature and
pressure, 69

Stannates, 917

Stannic compounds, 917 ; estima-
tion of, 920

Stannous compounds, 914 ; esti-
mation of, 920

Starch, 316, 695 ; iodide of, 407

Stassfurt potash deposits, 790

Steel, 979 ; cutting of, 189

Stephanite, 819

Stibine, 938

Stimmi, 932

Stoichiometry, 110

Stokes 's equation, 11

Stoneware, 903

Stream tin, 913

Stromeyerite, 819

Strontia, 852

Strontianite, 851

Strontium, 850 ; salts of, 851

Sublimation, 10, 76

Submicrons, 8

Substances, amphoteric, 863 ;
enantiotropic, 479 ; monotropic,
479; pure, 5, 25, 112

Suction gas, 705

Sugar, 695 ; of lead, 928

Sulphamide, 597

Sulphates, 512

Sulphides, 484

Sulphimide, 597

Sulphites, 495

Sulphur, 474 ; allot ropic forms
of, 478, 48»1 ; chlorides, 487;
colloidal, 482 ; combustion of,
490 ; compounds of, 490 ; di-
oxide, 490 ; flowers of, 475 ;
fluoride, 48-8 ; halogen com-
pounds of, 487 ; manufacture of,
474-477 ; oxygen compounds
of, 490 ; pure, 482 ; sesquioxide,
526 ; trioxide, 497 ; uses of,
477; vapour, 482

Sulphuretted hydrogen, 483

Sulphuryl chloride, 515 ; -group,
515

Sun, composition of, 32 ; energy of,
1035

Supersaturation, 101

INDEX

1061

Surface tension, 10, 270
Sylvanite, 532
Sylvine, 791

Symmetry of crystals, 434 ; ele-
ments of, 435
Sympathetic ink, 1000
Syngenite, 847
Synthesis, 26, 713

Talc, 746
Tantalite, 907
Tantalum, 944, 1007
Tartar emetic, 940
Tautomeric modifications, 717
Tellurium, 531, 832 ; atomic weight

of, 533 ; compounds of, 832, 833
Temperature, absolute, 67, 264 ;

critical, 170

Tempering of steel, 982
Tenorite, 805
Tetartohedral forms, 440
Tetrachromates, 954
Tetradymite, 532
Tetragonal system, 437
Tetrahedron, 441
Tetrakis hexahedron, 436
Tetrasilane, 749
Thallium, 890, 904 ; compounds of,

905

Thenard's blue, 897, 1000
Thermal constants, 707
Thermo-chemistry, 388
Thermo-couples, 1010
Thiazyl, chloride, 596 ; -nitrate,

596

Thioantimoniates, 938
Thioarsenates, 656
Thioarsenites, 656
Thiocarburyl chloride, 713
Thion hudor, 847
Thionyl, bromide, 496 ; chloride,

495 ; chlorobromide, 496 ; fluor-
ide, 496

Thiophosphoryl chloride, 637
Thiostannates, 919
Thomson's process for soda,

898

Thorianite, 930
Thorite, 930
Thorium, 930, 1028 ; compounds of,

930 ; disintegration series, 1029 ;

emanation, 1028 ; -X, 1028
Thulium, 1031
Tin, 912 ; alloys of, 915 ; black, 913 ;

dioxide, 914 ; metallurgy of,

913 ; plate, 914 ; rhombic form,

914 ; stream-, 913 ; white, 914

Tinning, 913
Tinstone, 912
Titaniferous iron ore, 929
Titanium, 929 ; compounds of, 929
Topaz, oriental, 894
Tourmaline, 890
Trapezohedron, 441
Trans-isomer, 1013
Transitional elements, 972
Transmutation of metals, 28
Trichloramine, 556
Trichromates, 954
Triclinic system, 440
Tridymite, 740, 742
Triethylsilicoformate, 748
Triakis octahedron, 436
Trinitrotoluene, 570
Triphylite, 795
Triple point, 92
Trisilane, 749
Trona, 785
Tungsten, 957 ; compounds of,

937

Turnbull's blue, 996
Turner's yellow, 925
Turquoise, 891
Tutia, 859
Twin crystals, 442

Ultramarine, 903
Ultra-microscope, 7

Uranium, 958 ; compounds of,
958 ; equivalent of, 430 ; radio-
activity of, 1027 ; series of
transformation of, 1028

Urao, 785

Urea, 708, 718

Vacuum vessels, 174, 178
Valency, 245, 1011 ; periodicity of,

463 ; volume, 246 ; residual,

253 ; supplementary, 1011
Valentine, Basil, 29, 932
Valentinite, 934
Vanadium, 944 ; atomic weight of,

445

Van der Waals' equation, 269
Van't Hoff' s theory of solutions,

311
Vapour, saturated and unsaturated,

74
Vapour densities, 81 ; abnormal,

150 ; Dumas' method, 83 ;

Hofmann's method, 81 ; Victor

Meyer's method, 86

1062

INDEX

Vapour pressure, -curve of water,
75 ; relative lowering of, 302 ;
of solids, 76 ; table of, 77

Varec, 403

Velocity constants, 352 ; of re-
action, 351 ; of sound, 263

Venetian white, 928

Verdigris, 814

Vinasse, 790

Vinegar, 771

Viscosity, 11

Vitrain, 669

Vitreosil, 743

Volhard's test for manganese,
924

Voltage, 880

Voltaic cell, 879

Voltameter, see Coulometer

Voltoids, 799

Volts, 880

Volume, atomic, 453 ; critical, 171 ;
Law of gaseous, 138 ; specific,
92

Wad, 961

Washing soda, 784, 785

Water, action of, on metals, 211 ;
aerated, 689 ; bacteriology of,
211 ; composition of, 51, 58, 60,
213, 216 ; of constitution, 813 ;
of crystallisation, 101 ; -gas,
683, 705, 728; hard and soft,
206, 209 ; ionisation of, 286,
362 ; mineral, 210 ; natural,
205 ; phases of, 91 ; physical
properties of, 200 ; -proofing,
896 ; pure, 212 ; rain, 2f)5 ;
river, 205, 209; spring, 210;
table of vapour pressures of,
77 ; -vapour in air, 79 ; vapour
pressure of, 203

Watson, Bishop R.9 103, 680, 1006

Watts, 880

Wave-length, 755 ; determination
of, by spectroscope, 762

Weathering, 694 ; of rocks, 789

Weight, gram molecular, 149

Welding, 189

Weldon process, 240

Welsbach incandescent gas man-
tles, 930

Welsh process, 806

Werner's theory of complex com-
pounds, 1010
Weston cell, 870
Wet process for silver, 928
White lead, 928
White nickel ore, 1002
White vitriol, 864
Willemite, 746
Williamson's violet, 995
Wilson, C. R. T., 1024
Witherite, 851
Wolfram, 957
Wollastonite, 746
Wollaston wires, 1007
Wood, distillation of, 605
Wood's fusible metal, 866

X-rays, 9, 756, 1017
Xenon, 603, 605
Xenophanes, 125
Xenotine, 907

Yellow prussiate of potash,
Ytterbium, 907
Yttrium, 907
Yttrotantalite, 907

Zaffre, 998
Zeolites, 789
Zeppelin, 190

Zinc, 859 ; ammonium compounds
865 ; arsenide, 648 ; aton.
weight of, 865 ; blende, 860

864 ; bromide, 864 ; carbon.
865; chloride, 863; chromat
955 ; cyanide, 865 ; dust, 861
estimation of, 865 ; ethyl, 86-^
foil, 862 ; granulated, 86
hydroxide, 863 ; iodide, 86*
metallurgy of, 860 ; mineral
859 ; nitrate, 865 ; nitri<
865; oxides, 863; phosph.-

865 ; sulphate, 185, 864 ; t.j1
phide, 864

Zincates, 863
Zincite, 860
Zinken, 860
Zircon, 746, 764, 929
Zirconium, 929 ; compounds, 929
Zosimus, 159

PRINTED IN GEEAT BRITAIN BT R. CLAT AND SONS, LTD.,
BRUNSWICK STREET, STAMFORD STREET, S.E. I, AND BUNGAT, SUFFOLK.

INTERNATIONAL ATOMIC WEIGHTS (1921).

Atomic

weight.

^ment. Syi

ml .,1. H - 1

O = 16

•\ minium .

\! -J-8

27-1

>" imony . .

F- 119-2

120-2

• on

39-9

*enic

As 74-37

74-96

'^ium

lj;. 136-28

137-37 -

jylliu'-

Be 9-0

91 *

triuth

Bi 206-4

208-0

Jron ...

B 10-8

10-9

kmine

Br 79-29

79-92

Sniium

Ccl 111-51

112-40

jsium

(* 131-76

132-81

•ciuni ....

Ca 39-75

40-07

Eon

C 11-910

12-005

hum

Co 139-15

140-25

lormo

Cl 35-18

35-46-

romhirn . .

Cr 51:6

52-0

halt

Co 58-50

58-97

pper

Cu 63-07

63-57

pprosium ..

Py 161-2

162-5

hi urn

Er 10G-4

167-7

Jropiuw

Eu 150-8

152-0

toriiiti . . . .

F 18-9

19-0

Qd 156-1

157:3

vli 1.111 1

Ga 69-5

70-1

mi . .

71-9

72-5

Id ....

Au 195-6

197-2

He 3-97

4-00

"Ho 162-2

163-5

H 1-000

1-008

.13-9

114-8

1 125-91

126-92

'/•6'

1Q3-1

••>• 40

J5-84

»vr 82- 2(»

82-92

I., i 13"'0

139-0

Pb 20

207 -2(j "

Li 6-89

6-94

La 173-6

175-0

1 1 a ...

Mg 24-13

24-32

^e ...

Mil 54-49

54-93

•cury

Hg 199-0

200-6

ybdenum

Mo 95- '2

96-0

Atomic weight.

Element. Symbol. H = 1

Neodymium... Nd 143-2

Neon Ne 20-0

Nickel Ni 58-21

Niobium Nb 92-4

Niton Nt 220-6

Nitrogen N 13-897

Osmium Os 189-4

Oxygen O 15-87

Palladium ... Pd 105-9

Phosphorus P 30-79

Platinum Pt 193-6

Potassium ... K 38-79
Praseodymium Pr 139-8

Radium Ra 224-2

Rhodium ..
Rubidium
Ruthenium
Samarium

Scandium
Selenium .

Rh 102-1
Rb 84-77
Ru 100-9
Sa 149-?
Sc 44-7
So 78-G

Silicon Si 28-1

Silver Ag 107 04

Na 22-S2
Sr 86-93
S 31-81
Ta 180-1
Te 126-5
Tb 157-9
Tl 202-4
Th 230-31
Tm 167-2
Sn 117-8
Ti 47-72
W 182-5
U 236-3
V 50-6

Xenon Xe 129-2

Ytterbium ... Yb 172-1

Yttrium Yt 88-62

Zinc Zn 64-85

Zirconium ... Zr 89-9

Sodium

Strontium . .

Sulphur

Tantalum
Tellurium

Terbium

Thallium

Thorium

Thulium

Tin

Titanium

Tungsten

Uranium ,

Vanadium . .

O = 16
144-3

20-2

58-68

93-1
222-4

14-008
190-9

16-00
106-7

31-04
195-2

39-10
140-9
226-0
102-9

85-45
101-7
1GO-4

45-1

79-2

28-3
107-88

23-00

87-63

32-06
181-5
127-5
159-2
204-0
232-15
168-5
118-7

48-1
134-0
238-2
51-0
130-2
173-5

80-33

65-37
90-6

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "A text-book of inorganic chemistry for university students"

See other formats